Surface parse errors

.github/workflows/build_package_ts.yml

@@ -19,6 +19,8 @@ jobs:
         uses: actions/checkout@v5
 
       - uses: pnpm/action-setup@v4
+        with:
+          version: 10
 
       - name: Setup Node.js
         uses: actions/setup-node@v4

.github/workflows/lint_py.yml

@@ -1,4 +1,4 @@
-name: Lint
+name: Lint - Python
 
 on:
   push:
@@ -29,18 +29,7 @@ jobs:
       - name: Set up Python
         run: uv python install ${{ matrix.python-version }}
 
-      - uses: pnpm/action-setup@v4
-      - name: Setup Node.js
-        uses: actions/setup-node@v4
-        with:
-          node-version-file: "ts/llama_cloud_services/.nvmrc"
-      - name: Install dependencies
-        run: pnpm install --no-frozen-lockfile
-
       - name: Run linter
         shell: bash
         working-directory: py
         run: uv run -- pre-commit run -a
-      # the js checks are run roundaboutly through lint-staged, and -a doesn't run it. Run them directly.
-      - run: pnpm -w --filter llama-cloud-services run lint
-      - run: pnpm -w --filter llama-cloud-services run format:check

.github/workflows/lint_ts.yml

@@ -0,0 +1,37 @@
+name: Lint - TypeScript
+
+on:
+  push:
+    branches:
+      - main
+    paths:
+      - "ts/**"
+  pull_request:
+    paths:
+      - "ts/**"
+
+env:
+  TURBO_TOKEN: ${{ secrets.TURBO_TOKEN }}
+  TURBO_TEAM: ${{ vars.TURBO_TEAM }}
+  TURBO_REMOTE_ONLY: true
+
+jobs:
+  lint:
+    runs-on: ubuntu-latest
+    steps:
+      - uses: actions/checkout@v5
+      - uses: pnpm/action-setup@v4
+        with:
+          version: 10
+      - name: Setup Node.js
+        uses: actions/setup-node@v4
+        with:
+          node-version-file: "ts/llama_cloud_services/.nvmrc"
+      - name: Install dependencies
+        run: pnpm install --no-frozen-lockfile
+      - name: Run lint
+        working-directory: ts/llama_cloud_services/
+        run: pnpm run lint
+      - name: Run Prettier
+        working-directory: ts/llama_cloud_services/
+        run: pnpm run format

.github/workflows/release_ts.yml

@@ -13,6 +13,8 @@ jobs:
         uses: actions/checkout@v5
 
       - uses: pnpm/action-setup@v4
+        with:
+          version: 10
 
       - name: Setup Node.js
         uses: actions/setup-node@v4
@@ -47,6 +49,6 @@ jobs:
         uses: ncipollo/release-action@v1
         with:
           artifacts: "ts/llama_cloud_services/llama-cloud-services*.tgz"
-          name: Release ${{ github.ref_name }} - LlamaCloud Services TS
-          generateReleaseNotes: true
+          name: Release ${{ github.ref }} - LlamaCloud Services TS
+          bodyFile: "ts/llama_cloud_services/CHANGELOG.md"
           token: ${{ secrets.GITHUB_TOKEN }}

.github/workflows/test_ts.yml

@@ -1,4 +1,4 @@
-name: Test - TypeScript
+name: Lint - TypeScript
 
 on:
   push:
@@ -23,14 +23,17 @@ jobs:
     steps:
       - uses: actions/checkout@v5
       - uses: pnpm/action-setup@v4
+        with:
+          version: 10
       - name: Setup Node.js
         uses: actions/setup-node@v4
         with:
           node-version-file: "ts/llama_cloud_services/.nvmrc"
       - name: Install dependencies
-        run: pnpm -r install --no-frozen-lockfile
-      - name: Build package
-        run: pnpm --filter llama-cloud-services build
+        run: pnpm install --no-frozen-lockfile
+      - name: Run Build
+        working-directory: ts/llama_cloud_services/
+        run: pnpm build
       - name: Run Tests
         working-directory: ts/llama_cloud_services/
         run: pnpm test

.pre-commit-config.yaml

@@ -60,13 +60,11 @@ repos:
         additional_dependencies: [black==23.10.1]
         # Using PEP 8's line length in docs prevents excess left/right scrolling
         args: [--line-length=79]
-  - repo: local
+  - repo: https://github.com/pre-commit/mirrors-prettier
+    rev: v3.0.3
     hooks:
-      - id: lint-staged
-        name: Run lint-staged for TS files
-        entry: pnpm -w exec lint-staged
-        language: system
-        pass_filenames: false
+      - id: prettier
+        exclude: ^(uv.lock|ts/llama_cloud_services/pnpm-lock.yaml|ts/e2e-tests)
   - repo: https://github.com/codespell-project/codespell
     rev: v2.2.6
     hooks:

CONTRIBUTING.md

@@ -18,7 +18,7 @@ versions need to be kept consistent to sidecar it with `llama_cloud_services`. B
 
 You can also do this with `./scripts/version-bump.py set 0.x.x` if you have `uv` installed.
 
-Once the change is merged, push a tag `git tag -a v0.x.x -m 0.x.x` and `git push origin v0.x.x`.
+Once the change is merged, push a tag `git tag -a v0.x.x -m 0.x.x` and `git push origin 0.x.x`.
 
 This tagging step can be done with `./scripts/version-bump tag`.
 

README.md

@@ -9,6 +9,7 @@ This repository contains the code for hand-written SDKs and clients for interact
 This includes:
 
 - [LlamaParse](./parse.md) - A GenAI-native document parser that can parse complex document data for any downstream LLM use case (Agents, RAG, data processing, etc.).
+- [LlamaReport (beta/invite-only)](./report.md) - A prebuilt agentic report builder that can be used to build reports from a variety of data sources.
 - [LlamaExtract](./extract.md) - A prebuilt agentic data extractor that can be used to transform data into a structured JSON representation.
 - [LlamaCloud Index](./index.md) - A widely customizable and fully automated document ingestion pipeline that also serves retrieval purposes.
 
@@ -27,11 +28,13 @@ Then, you can use the services in your code:
 ```python
 from llama_cloud_services import (
     LlamaParse,
+    LlamaReport,
     LlamaExtract,
     LlamaCloudIndex,
 )
 
 parser = LlamaParse(api_key="YOUR_API_KEY")
+report = LlamaReport(api_key="YOUR_API_KEY")
 extract = LlamaExtract(api_key="YOUR_API_KEY")
 index = LlamaCloudIndex(
     "my_first_index", project_name="default", api_key="YOUR_API_KEY"
@@ -41,6 +44,7 @@ index = LlamaCloudIndex(
 See the quickstart guides for each service for more information:
 
 - [LlamaParse](./parse.md)
+- [LlamaReport (beta/invite-only)](./report.md)
 - [LlamaExtract](./extract.md)
 - [LlamaCloud Index](./index.md)
 
@@ -53,11 +57,13 @@ You can also create your API key in the EU region [here](https://cloud.eu.llamai
 ```python
 from llama_cloud_services import (
     LlamaParse,
+    LlamaReport,
     LlamaExtract,
     EU_BASE_URL,
 )
 
 parser = LlamaParse(api_key="YOUR_API_KEY", base_url=EU_BASE_URL)
+report = LlamaReport(api_key="YOUR_API_KEY", base_url=EU_BASE_URL)
 extract = LlamaExtract(api_key="YOUR_API_KEY", base_url=EU_BASE_URL)
 index = LlamaCloudIndex(
     "my_first_index",

examples/README.md

@@ -4,6 +4,6 @@ In this folder you will find several python notebooks that contain examples rega
 
 - [LlamaParse](./parse/)
 - [LlamaExtract](./extract/)
-- [LlamaCloudIndex](./index/)
+- [LlamaReport](./report/)
 
 Follow the instructions in each notebook to get started!

examples/parse/demo_excel.ipynb

@@ -43,7 +43,7 @@
    "source": [
     "from llama_cloud_services import LlamaParse\n",
     "\n",
-    "api_key = \"llx-...\"  # get from cloud.llamaindex.ai"
+    "api_key = \"llx-jwAQZL8T38onyL9hKBOXyRtnuCU0Fk3z7tmDhIT3L0GEfohJ\"  # get from cloud.llamaindex.ai"
    ]
   },
   {

examples/parse/excel/dcf_rag.ipynb

@@ -75,6 +75,7 @@
     "    adaptive_long_table=True,\n",
     "    outlined_table_extraction=True,\n",
     "    output_tables_as_HTML=True,\n",
+    "    api_key=\"llx-jwAQZL8T38onyL9hKBOXyRtnuCU0Fk3z7tmDhIT3L0GEfohJ\",\n",
     ")\n",
     "\n",
     "result = await parser.aparse(\"./dcf_template.xlsx\")\n",

examples/report/basic_report.ipynb

@@ -0,0 +1,762 @@
+{
+ "cells": [
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "# Report Generation with LlamaReport\n",
+    "\n",
+    "In this notebook, we'll walk through the basic process of generating a report with LlamaReport, and highlight some of the key features of the library.\n",
+    "\n",
+    "TLDR:\n",
+    "1. Download source data to use as knowledge base for the report\n",
+    "2. Kick off report generation with a template\n",
+    "3. Get the plan and review/accept/reject suggestions\n",
+    "4. Get the final report\n",
+    "5. Review/accept/reject suggestions to edit the final report\n",
+    "6. Print the final report"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "%pip install llama-cloud-services"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## 1. Download Source Data\n",
+    "\n",
+    "Here, we download the `Attention is All You Need` paper as a PDF.\n",
+    "\n",
+    "LlamaReport currently supports up to 5 files as input, and essentially any file type that can be parsed by LlamaParse.\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "!wget \"https://arxiv.org/pdf/1706.03762.pdf\" -O \"./attention.pdf\""
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## 2. Kick off Report Generation\n",
+    "\n",
+    "Here, we kick off report generation with a template.\n",
+    "\n",
+    "The template can either be a string or a file path, but here we'll use a string.\n",
+    "\n",
+    "In our experiments, anything works as a template, but some general guidelines:\n",
+    "\n",
+    "- Use markdown formatting + instructions in each section to guide the report generation\n",
+    "- If using an existing file as a template, provide extra instructions to guide the report generation\n",
+    "\n",
+    "**NOTE:** Since we are in a notebook, we will use async functions and `await` throughout. Synchronous methods that work without `await` are available by just removing the `a` from the method name and removing the `await` keyword."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "from llama_cloud_services import LlamaReport\n",
+    "\n",
+    "llama_report = LlamaReport(\n",
+    "    api_key=\"llx-...\",\n",
+    ")\n",
+    "\n",
+    "report_client = await llama_report.acreate_report(\n",
+    "    name=\"my_cool_report_on_attention\",\n",
+    "    # can pass in file paths or bytes\n",
+    "    input_files=[\"./attention.pdf\"],\n",
+    "    template_text=\"\"\"\\\n",
+    "# [Some title]\\n\\n\n",
+    "## TLDR\\n\n",
+    "A quick summary of the paper.\\n\\n\n",
+    "## Details\\n\n",
+    "More details about the paper, possibly more than one section here.\\n\n",
+    "\"\"\",\n",
+    "    # optional additional instructions for the report generation\n",
+    "    # template_instructions=None,\n",
+    "    # optional file path to an existing template instead of template_text\n",
+    "    # template_file=None,\n",
+    ")"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "The returned `ReportClient` object is used to interact with the report generation process for this specific report."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Report(id=0a394b33-1a3e-463c-b5cb-7ff8ab827d0a, name=my_cool_report_on_attention)\n"
+     ]
+    }
+   ],
+   "source": [
+    "print(report_client)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## 3. Get the plan\n",
+    "\n",
+    "The first phases of report generation involve ingesting the source data and generating a plan.\n",
+    "\n",
+    "The plan is a list of instructions for the report generation, and can be reviewed/accepted/rejected by the user.\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "plan = await report_client.await_for_plan(\n",
+    "    timeout=10000,\n",
+    "    poll_interval=10,\n",
+    ")"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "# {title}\n",
+      "[ReportQuery(field='title', prompt='Generate a clear and concise title for this paper about the Transformer model and attention mechanisms', context='The paper discusses the Transformer architecture for sequence transduction using attention mechanisms, focusing on machine translation applications')]\n",
+      "==================\n",
+      "## TLDR\n",
+      "\n",
+      "{tldr_content}\n",
+      "[ReportQuery(field='tldr_content', prompt='Write a brief, clear summary of the key points about the Transformer model', context='Focus on the main innovations: attention mechanisms, efficiency improvements, and state-of-the-art results in machine translation')]\n",
+      "==================\n",
+      "## Details\n",
+      "\n",
+      "{details_content}\n",
+      "[ReportQuery(field='details_content', prompt='Provide detailed information about the Transformer model architecture and its applications', context='Include information about:\\n- The attention mechanism implementation\\n- Advantages over recurrent and convolutional models\\n- Performance in machine translation tasks\\n- Training efficiency improvements')]\n",
+      "==================\n"
+     ]
+    }
+   ],
+   "source": [
+    "for plan_block in plan.blocks:\n",
+    "    print(plan_block.block.template)\n",
+    "    print(plan_block.queries)\n",
+    "    print(\"==================\")"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "With the plan, we can either use it to kick off generation of the final report, or we can edit the plan and adjust it as needed.\n",
+    "\n",
+    "While we could manually edit the objects here and use `await report_client.aupdate_plan(action=\"edit\", updated_plan=plan)`, we can also use `LlamaReport` to agentically edit the plan."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "suggestions = await report_client.asuggest_edits(\n",
+    "    \"Can you split the details section into two sections?\"\n",
+    ")"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Justification for change: \n",
+      "I'll help you break down the details section into two distinct parts - one focusing on the architecture and another on the practical applications and performance. This will make the content more organized and easier to follow. The original block at index 2 will be replaced with these two new sections.\n",
+      "\n",
+      "Proposed changes:\n",
+      "\n",
+      "## Architecture Details\n",
+      "\n",
+      "{architecture_content}\n",
+      "\n",
+      "[ReportQuery(field='architecture_content', prompt='Describe the technical details of the Transformer model architecture', context='Focus on:\\n- Core components of the Transformer architecture\\n- Self-attention mechanism implementation\\n- Multi-head attention details\\n- Position encoding approach\\n- Feed-forward network structure')]\n",
+      "==================\n",
+      "\n",
+      "## Performance and Applications\n",
+      "\n",
+      "{applications_content}\n",
+      "\n",
+      "[ReportQuery(field='applications_content', prompt='Explain the practical applications and performance advantages of the Transformer model', context='Cover:\\n- Comparison with RNN and CNN models\\n- Machine translation results and benchmarks\\n- Training efficiency improvements\\n- Real-world applications and use cases\\n- Scalability benefits')]\n",
+      "==================\n"
+     ]
+    }
+   ],
+   "source": [
+    "for suggestion in suggestions:\n",
+    "    print(\"Justification for change:\", suggestion.justification)\n",
+    "    print(\"Proposed changes:\")\n",
+    "    for plan_block in suggestion.blocks:\n",
+    "        print(plan_block.block.template)\n",
+    "        print(plan_block.queries)\n",
+    "        print(\"==================\")"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "This looks pretty good! We can also use the client to automatically accept and apply, or reject, these suggestions.\n",
+    "\n",
+    "This will (locally) keep track of the history of changes, so that future suggestions can be based on the previous changes."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "for suggestion in suggestions:\n",
+    "    await report_client.aaccept_edit(suggestion)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "What effect did that have on the tracked local history? Let's see!"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "[EditAction(block_idx=2, old_content='## Details\\n\\n{details_content}\\n\\nField: details_content, Prompt: Provide detailed information about the Transformer model architecture and its applications, Context: Include information about:\\n- The attention mechanism implementation\\n- Advantages over recurrent and convolutional models\\n- Performance in machine translation tasks\\n- Training efficiency improvements\\nDepends on: none', new_content='\\n## Architecture Details\\n\\n{architecture_content}\\n\\n\\nField: architecture_content, Prompt: Describe the technical details of the Transformer model architecture, Context: Focus on:\\n- Core components of the Transformer architecture\\n- Self-attention mechanism implementation\\n- Multi-head attention details\\n- Position encoding approach\\n- Feed-forward network structure\\nDepends on: none', action='approved', timestamp=datetime.datetime(2025, 2, 4, 20, 59, 55, 773558)),\n",
+       " EditAction(block_idx=3, old_content='[No old content]', new_content='\\n## Performance and Applications\\n\\n{applications_content}\\n\\n\\nField: applications_content, Prompt: Explain the practical applications and performance advantages of the Transformer model, Context: Cover:\\n- Comparison with RNN and CNN models\\n- Machine translation results and benchmarks\\n- Training efficiency improvements\\n- Real-world applications and use cases\\n- Scalability benefits\\nDepends on: previous', action='approved', timestamp=datetime.datetime(2025, 2, 4, 20, 59, 55, 773687))]"
+      ]
+     },
+     "execution_count": null,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "report_client.edit_history"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "[Message(role=<MessageRole.USER: 'user'>, content='Can you split the details section into two sections?', timestamp=datetime.datetime(2025, 2, 4, 20, 59, 47, 754848)),\n",
+       " Message(role=<MessageRole.ASSISTANT: 'assistant'>, content=\"\\nI'll help you break down the details section into two distinct parts - one focusing on the architecture and another on the practical applications and performance. This will make the content more organized and easier to follow. The original block at index 2 will be replaced with these two new sections.\\n\", timestamp=datetime.datetime(2025, 2, 4, 20, 59, 55, 482070))]"
+      ]
+     },
+     "execution_count": null,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "report_client.chat_history"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "These two items are used to provide context for future suggestions! You can always clear this, or provide your own history."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# report_client.suggest_edits(\"....\", chat_history=[{\"role\": \"user\", \"content\": \"...\"}, ...])"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## 4. Get the final report\n",
+    "\n",
+    "Now that we have a plan, we can kick off generation of the final report."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# kicks off report generation\n",
+    "await report_client.aupdate_plan(action=\"approve\")\n",
+    "\n",
+    "# waits for report generation to complete\n",
+    "report = await report_client.await_completion(\n",
+    "    timeout=10000,\n",
+    "    poll_interval=10,\n",
+    ")"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "# Attention Is All You Need: A Pure Attention-Based Architecture for Neural Machine Translation\n",
+      "\n",
+      "## TLDR\n",
+      "\n",
+      "The Transformer introduced a revolutionary architecture that relies entirely on attention mechanisms, eliminating the need for recurrence or convolution in sequence processing. Its key innovations include multi-head self-attention for parallel processing of input sequences, scaled dot-product attention for efficient computation, and positional encodings for sequence order awareness. The model achieved breakthrough results in machine translation (28.4 BLEU on English-to-German, 41.8 BLEU on English-to-French) while requiring significantly less training time than previous approaches, training in 3.5 days on 8 GPUs. This architecture demonstrated that attention mechanisms alone are sufficient for state-of-the-art sequence modeling, setting a new direction for natural language processing.\n",
+      "\n",
+      "\n",
+      "## Architecture Details\n",
+      "\n",
+      "The Transformer architecture represents a groundbreaking approach to sequence processing, built entirely on attention mechanisms without recurrence or convolution. Here are its key technical details:\n",
+      "\n",
+      "Core Components:\n",
+      "- Encoder-decoder architecture with stacked self-attention and point-wise feed-forward layers\n",
+      "- Each layer contains two main sub-layers: multi-head self-attention mechanism and position-wise feed-forward network\n",
+      "- Layer normalization and residual connections between sub-layers\n",
+      "- No recurrent or convolutional elements, enabling parallel processing\n",
+      "\n",
+      "Self-Attention Mechanism:\n",
+      "- Processes relationships between all positions in a sequence simultaneously\n",
+      "- Computes attention weights using queries, keys, and values derived from input representations\n",
+      "- Implements scaled dot-product attention to prevent gradient issues with large input dimensions\n",
+      "- Allows direct modeling of dependencies regardless of positional distance\n",
+      "- Uses masking in decoder to prevent leftward information flow and maintain auto-regressive property\n",
+      "\n",
+      "Multi-Head Attention:\n",
+      "- Employs multiple attention heads operating in parallel\n",
+      "- Each head processes information in different representation subspaces\n",
+      "- Three types of attention applications:\n",
+      "  1. Encoder self-attention (all positions attend to each other)\n",
+      "  2. Decoder self-attention (each position attends to previous positions)\n",
+      "  3. Encoder-decoder attention (decoder queries attend to encoder outputs)\n",
+      "- Counteracts reduced resolution from attention averaging through parallel processing\n",
+      "\n",
+      "Position-wise Feed-Forward Network:\n",
+      "- Applied identically to each position separately\n",
+      "- Consists of two linear transformations with ReLU activation\n",
+      "- Structure: FFN(x) = max(0, xW1 + b1)W2 + b2\n",
+      "- Input and output dimensionality: dmodel = 512\n",
+      "- Inner-layer dimensionality: dff = 2048\n",
+      "- Parameters vary between layers but remain constant across positions\n",
+      "\n",
+      "Position Encoding:\n",
+      "- Adds positional information to input embeddings\n",
+      "- Enables the model to consider sequential order without recurrence\n",
+      "- Implements sinusoidal position encodings to allow model to attend to relative positions\n",
+      "- Maintains constant number of operations between any two positions, unlike convolutional approaches\n",
+      "- Allows effective modeling of both local and long-range dependencies\n",
+      "\n",
+      "\n",
+      "\n",
+      "## Performance and Applications\n",
+      "\n",
+      "The Transformer model demonstrates significant performance advantages and practical applications across multiple domains:\n",
+      "\n",
+      "Performance Advantages over RNN/CNN Models:\n",
+      "- Eliminates sequential computation constraints present in RNNs, enabling superior parallelization\n",
+      "- Reduces operations needed for relating distant positions to a constant number, compared to linear/logarithmic scaling in CNNs\n",
+      "- Processes all input and output positions simultaneously through self-attention mechanisms\n",
+      "- Achieves state-of-the-art results while requiring significantly less computational resources\n",
+      "\n",
+      "Machine Translation Benchmarks:\n",
+      "- WMT 2014 English-to-German: 28.4 BLEU score, exceeding previous best results by over 2 BLEU points\n",
+      "- WMT 2014 English-to-French: 41.8 BLEU score (single-model state-of-the-art)\n",
+      "- Surpasses performance of existing model ensembles in translation tasks\n",
+      "\n",
+      "Training Efficiency:\n",
+      "- Requires only 3.5 days of training on eight GPUs for state-of-the-art performance\n",
+      "- Achieves superior results at \"a small fraction of the training costs\" compared to previous models\n",
+      "- Enables significantly faster training through parallel processing of input/output sequences\n",
+      "- Can reach production-quality performance in as little as twelve hours on modern GPU hardware\n",
+      "\n",
+      "Real-world Applications:\n",
+      "- Machine translation systems\n",
+      "- Natural language understanding tasks\n",
+      "- Reading comprehension\n",
+      "- Abstractive summarization\n",
+      "- Text entailment analysis\n",
+      "- Constituency parsing (achieving 92.7 F1 score in semi-supervised settings)\n",
+      "- Adaptable to both large and limited training data scenarios\n",
+      "\n",
+      "Scalability Benefits:\n",
+      "- Highly parallelizable architecture enables efficient scaling across multiple GPUs\n",
+      "- Constant computational complexity for relating any input/output positions\n",
+      "- Effective handling of long-range dependencies in sequences\n",
+      "- Maintains performance quality while scaling to larger datasets and model sizes\n",
+      "- Generalizes well across different tasks and domains without architectural changes\n",
+      "- Supports efficient inference and deployment in production environments\n",
+      "\n"
+     ]
+    }
+   ],
+   "source": [
+    "report_text = \"\\n\\n\".join([block.template for block in report.blocks])\n",
+    "print(report_text)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## 5. Edit the final report\n",
+    "\n",
+    "Now that we have a report, we can edit it.\n",
+    "\n",
+    "We can use the `asuggest_edits` method to get suggestions for edits, and then use the `aaccept_edit`/`areject_edit` methods to apply them.\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Justification for change: \n",
+      "I'd suggest changing \"TLDR\" to \"Executive Summary\" which is more appropriate for a professional or academic report. This term is widely used in formal documents and better reflects the nature of this concise overview section while maintaining the same function of providing a quick summary of the key points.\n",
+      "\n",
+      "Proposed changes:\n",
+      "## Executive Summary\n",
+      "\n",
+      "The Transformer introduced a revolutionary architecture that relies entirely on attention mechanisms, eliminating the need for recurrence or convolution in sequence processing. Its key innovations include multi-head self-attention for parallel processing of input sequences, scaled dot-product attention for efficient computation, and positional encodings for sequence order awareness. The model achieved breakthrough results in machine translation (28.4 BLEU on English-to-German, 41.8 BLEU on English-to-French) while requiring significantly less training time than previous approaches, training in 3.5 days on 8 GPUs. This architecture demonstrated that attention mechanisms alone are sufficient for state-of-the-art sequence modeling, setting a new direction for natural language processing.\n",
+      "==================\n"
+     ]
+    }
+   ],
+   "source": [
+    "suggestions = await report_client.asuggest_edits(\n",
+    "    \"Can you change the TLDR header to something more professional?\"\n",
+    ")\n",
+    "for suggestion in suggestions:\n",
+    "    print(\"Justification for change:\", suggestion.justification)\n",
+    "    print(\"Proposed changes:\")\n",
+    "    for block in suggestion.blocks:\n",
+    "        print(block.template)\n",
+    "        print(\"==================\")"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "Changing to \"Executive Summary\" sounds reasonable, lets accept that!\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "for suggestion in suggestions:\n",
+    "    await report_client.aaccept_edit(suggestion)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## 7. Print the final report\n",
+    "\n",
+    "Now that we have a report, we can print it."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "# Attention Is All You Need: A Pure Attention-Based Architecture for Neural Machine Translation\n",
+      "\n",
+      "## Executive Summary\n",
+      "\n",
+      "The Transformer introduced a revolutionary architecture that relies entirely on attention mechanisms, eliminating the need for recurrence or convolution in sequence processing. Its key innovations include multi-head self-attention for parallel processing of input sequences, scaled dot-product attention for efficient computation, and positional encodings for sequence order awareness. The model achieved breakthrough results in machine translation (28.4 BLEU on English-to-German, 41.8 BLEU on English-to-French) while requiring significantly less training time than previous approaches, training in 3.5 days on 8 GPUs. This architecture demonstrated that attention mechanisms alone are sufficient for state-of-the-art sequence modeling, setting a new direction for natural language processing.\n",
+      "\n",
+      "\n",
+      "## Architecture Details\n",
+      "\n",
+      "The Transformer architecture represents a groundbreaking approach to sequence processing, built entirely on attention mechanisms without recurrence or convolution. Here are its key technical details:\n",
+      "\n",
+      "Core Components:\n",
+      "- Encoder-decoder architecture with stacked self-attention and point-wise feed-forward layers\n",
+      "- Each layer contains two main sub-layers: multi-head self-attention mechanism and position-wise feed-forward network\n",
+      "- Layer normalization and residual connections between sub-layers\n",
+      "- No recurrent or convolutional elements, enabling parallel processing\n",
+      "\n",
+      "Self-Attention Mechanism:\n",
+      "- Processes relationships between all positions in a sequence simultaneously\n",
+      "- Computes attention weights using queries, keys, and values derived from input representations\n",
+      "- Implements scaled dot-product attention to prevent gradient issues with large input dimensions\n",
+      "- Allows direct modeling of dependencies regardless of positional distance\n",
+      "- Uses masking in decoder to prevent leftward information flow and maintain auto-regressive property\n",
+      "\n",
+      "Multi-Head Attention:\n",
+      "- Employs multiple attention heads operating in parallel\n",
+      "- Each head processes information in different representation subspaces\n",
+      "- Three types of attention applications:\n",
+      "  1. Encoder self-attention (all positions attend to each other)\n",
+      "  2. Decoder self-attention (each position attends to previous positions)\n",
+      "  3. Encoder-decoder attention (decoder queries attend to encoder outputs)\n",
+      "- Counteracts reduced resolution from attention averaging through parallel processing\n",
+      "\n",
+      "Position-wise Feed-Forward Network:\n",
+      "- Applied identically to each position separately\n",
+      "- Consists of two linear transformations with ReLU activation\n",
+      "- Structure: FFN(x) = max(0, xW1 + b1)W2 + b2\n",
+      "- Input and output dimensionality: dmodel = 512\n",
+      "- Inner-layer dimensionality: dff = 2048\n",
+      "- Parameters vary between layers but remain constant across positions\n",
+      "\n",
+      "Position Encoding:\n",
+      "- Adds positional information to input embeddings\n",
+      "- Enables the model to consider sequential order without recurrence\n",
+      "- Implements sinusoidal position encodings to allow model to attend to relative positions\n",
+      "- Maintains constant number of operations between any two positions, unlike convolutional approaches\n",
+      "- Allows effective modeling of both local and long-range dependencies\n",
+      "\n",
+      "\n",
+      "\n",
+      "## Performance and Applications\n",
+      "\n",
+      "The Transformer model demonstrates significant performance advantages and practical applications across multiple domains:\n",
+      "\n",
+      "Performance Advantages over RNN/CNN Models:\n",
+      "- Eliminates sequential computation constraints present in RNNs, enabling superior parallelization\n",
+      "- Reduces operations needed for relating distant positions to a constant number, compared to linear/logarithmic scaling in CNNs\n",
+      "- Processes all input and output positions simultaneously through self-attention mechanisms\n",
+      "- Achieves state-of-the-art results while requiring significantly less computational resources\n",
+      "\n",
+      "Machine Translation Benchmarks:\n",
+      "- WMT 2014 English-to-German: 28.4 BLEU score, exceeding previous best results by over 2 BLEU points\n",
+      "- WMT 2014 English-to-French: 41.8 BLEU score (single-model state-of-the-art)\n",
+      "- Surpasses performance of existing model ensembles in translation tasks\n",
+      "\n",
+      "Training Efficiency:\n",
+      "- Requires only 3.5 days of training on eight GPUs for state-of-the-art performance\n",
+      "- Achieves superior results at \"a small fraction of the training costs\" compared to previous models\n",
+      "- Enables significantly faster training through parallel processing of input/output sequences\n",
+      "- Can reach production-quality performance in as little as twelve hours on modern GPU hardware\n",
+      "\n",
+      "Real-world Applications:\n",
+      "- Machine translation systems\n",
+      "- Natural language understanding tasks\n",
+      "- Reading comprehension\n",
+      "- Abstractive summarization\n",
+      "- Text entailment analysis\n",
+      "- Constituency parsing (achieving 92.7 F1 score in semi-supervised settings)\n",
+      "- Adaptable to both large and limited training data scenarios\n",
+      "\n",
+      "Scalability Benefits:\n",
+      "- Highly parallelizable architecture enables efficient scaling across multiple GPUs\n",
+      "- Constant computational complexity for relating any input/output positions\n",
+      "- Effective handling of long-range dependencies in sequences\n",
+      "- Maintains performance quality while scaling to larger datasets and model sizes\n",
+      "- Generalizes well across different tasks and domains without architectural changes\n",
+      "- Supports efficient inference and deployment in production environments\n",
+      "\n"
+     ]
+    }
+   ],
+   "source": [
+    "report_response = await report_client.aget()\n",
+    "report_text = \"\\n\\n\".join([block.template for block in report_response.report.blocks])\n",
+    "print(report_text)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "We can also see the sources for each block!"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "0.99687636\n",
+      "# Abstract\n",
+      "\n",
+      "The dominant sequence transduction models are based on complex recurrent or convolutiona\n",
+      "==================\n",
+      "0.99591404\n",
+      "# 2 Background\n",
+      "\n",
+      "The goal of reducing sequential computation also forms the foundation of the Extende\n",
+      "==================\n",
+      "0.9951325\n",
+      "# 1 Introduction\n",
+      "\n",
+      "Recurrent neural networks, long short-term memory [13] and gated recurrent [7] neu\n",
+      "==================\n",
+      "0.99442345\n",
+      "# 7 Conclusion\n",
+      "\n",
+      "In this work, we presented the Transformer, the first sequence transduction model ba\n",
+      "==================\n",
+      "0.9967649\n",
+      "# 3.2.3 Applications of Attention in our Model\n",
+      "\n",
+      "The Transformer uses multi-head attention in three d\n",
+      "==================\n",
+      "0.99533635\n",
+      "# 2 Background\n",
+      "\n",
+      "The goal of reducing sequential computation also forms the foundation of the Extende\n",
+      "==================\n",
+      "0.9935868\n",
+      "# Abstract\n",
+      "\n",
+      "The dominant sequence transduction models are based on complex recurrent or convolutiona\n",
+      "==================\n",
+      "0.98780584\n",
+      "# Outputs\n",
+      "\n",
+      "(shifted right)\n",
+      "\n",
+      "Figure 1: The Transformer - model architecture.\n",
+      "\n",
+      "The Transformer follows\n",
+      "==================\n",
+      "0.9205043\n",
+      "# 3.3 Position-wise Feed-Forward Networks\n",
+      "\n",
+      "In addition to attention sub-layers, each of the layers i\n",
+      "==================\n",
+      "0.79581684\n",
+      "# 1 Introduction\n",
+      "\n",
+      "Recurrent neural networks, long short-term memory [13] and gated recurrent [7] neu\n",
+      "==================\n",
+      "0.9946774\n",
+      "# Abstract\n",
+      "\n",
+      "The dominant sequence transduction models are based on complex recurrent or convolutiona\n",
+      "==================\n",
+      "0.97079873\n",
+      "# 7 Conclusion\n",
+      "\n",
+      "In this work, we presented the Transformer, the first sequence transduction model ba\n",
+      "==================\n",
+      "0.9535353\n",
+      "# 6.3 English Constituency Parsing\n",
+      "\n",
+      "To evaluate if the Transformer can generalize to other tasks we \n",
+      "==================\n",
+      "0.9514138\n",
+      "# 2 Background\n",
+      "\n",
+      "The goal of reducing sequential computation also forms the foundation of the Extende\n",
+      "==================\n",
+      "0.9790758\n",
+      "# 1 Introduction\n",
+      "\n",
+      "Recurrent neural networks, long short-term memory [13] and gated recurrent [7] neu\n",
+      "==================\n",
+      "0.92262185\n",
+      "# Outputs\n",
+      "\n",
+      "(shifted right)\n",
+      "\n",
+      "Figure 1: The Transformer - model architecture.\n",
+      "\n",
+      "The Transformer follows\n",
+      "==================\n"
+     ]
+    }
+   ],
+   "source": [
+    "for block in report_response.report.blocks:\n",
+    "    # Each block has a list of sources, which are the nodes that were used to generate the block\n",
+    "    for source in block.sources:\n",
+    "        print(source.score)\n",
+    "        print(source.node.text[:100])\n",
+    "        print(\"==================\")"
+   ]
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "llama-parse-aNC435Vv-py3.10",
+   "language": "python",
+   "name": "python3"
+  },
+  "language_info": {
+   "codemirror_mode": {
+    "name": "ipython",
+    "version": 3
+   },
+   "file_extension": ".py",
+   "mimetype": "text/x-python",
+   "name": "python",
+   "nbconvert_exporter": "python",
+   "pygments_lexer": "ipython3"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 2
+}

package.json

@@ -1,19 +0,0 @@
-{
-  "name": "llama-cloud-services-workspace",
-  "version": "0.0.1",
-  "description": "",
-  "private": true,
-  "keywords": [],
-  "author": "",
-  "devDependencies": {
-    "prettier": "^3.6.2",
-    "lint-staged": "^15.4.2"
-  },
-  "lint-staged": {
-    "ts/llama_cloud_services/src/**/*.{ts,tsx,js,jsx}": [
-      "pnpm --filter llama-cloud-services exec eslint --fix",
-      "pnpm --filter llama-cloud-services exec prettier --write"
-    ]
-  },
-  "packageManager": "pnpm@10.11.1+sha512.e519b9f7639869dc8d5c3c5dfef73b3f091094b0a006d7317353c72b124e80e1afd429732e28705ad6bfa1ee879c1fce46c128ccebd3192101f43dd67c667912"
-}

pnpm-lock.yaml

@@ -6,15 +6,6 @@ settings:
 
 importers:
 
-  .:
-    devDependencies:
-      lint-staged:
-        specifier: ^15.4.2
-        version: 15.5.2
-      prettier:
-        specifier: ^3.6.2
-        version: 3.6.2
-
   ts/e2e-tests:
     devDependencies:
       '@types/node':
@@ -838,10 +829,6 @@ packages:
   ajv@8.17.1:
     resolution: {integrity: sha512-B/gBuNg5SiMTrPkC+A2+cW0RszwxYmn6VYxB/inlBStS5nx6xHIt/ehKRhIMhqusl7a8LjQoZnjCs5vhwxOQ1g==}
 
-  ansi-escapes@7.0.0:
-    resolution: {integrity: sha512-GdYO7a61mR0fOlAsvC9/rIHf7L96sBc6dEWzeOu+KAea5bZyQRPIpojrVoI4AXGJS/ycu/fBTdLrUkA4ODrvjw==}
-    engines: {node: '>=18'}
-
   ansi-regex@5.0.1:
     resolution: {integrity: sha512-quJQXlTSUGL2LH9SUXo8VwsY4soanhgo6LNSm84E1LBcE8s3O0wpdiRzyR9z/ZZJMlMWv37qOOb9pdJlMUEKFQ==}
     engines: {node: '>=8'}
@@ -946,10 +933,6 @@ packages:
     resolution: {integrity: sha512-ywqV+5MmyL4E7ybXgKys4DugZbX0FC6LnwrhjuykIjnK9k8OQacQ7axGKnjDXWNhns0xot3bZI5h55H8yo9cJg==}
     engines: {node: '>=6'}
 
-  cli-truncate@4.0.0:
-    resolution: {integrity: sha512-nPdaFdQ0h/GEigbPClz11D0v/ZJEwxmeVZGeMo3Z5StPtUTkA9o1lD6QwoirYiSDzbcwn2XcjwmCp68W1IS4TA==}
-    engines: {node: '>=18'}
-
   cliui@8.0.1:
     resolution: {integrity: sha512-BSeNnyus75C4//NQ9gQt1/csTXyo/8Sb+afLAkzAptFuMsod9HFokGNudZpi/oQV73hnVK+sR+5PVRMd+Dr7YQ==}
     engines: {node: '>=12'}
@@ -961,17 +944,10 @@ packages:
   color-name@1.1.4:
     resolution: {integrity: sha512-dOy+3AuW3a2wNbZHIuMZpTcgjGuLU/uBL/ubcZF9OXbDo8ff4O8yVp5Bf0efS8uEoYo5q4Fx7dY9OgQGXgAsQA==}
 
-  colorette@2.0.20:
-    resolution: {integrity: sha512-IfEDxwoWIjkeXL1eXcDiow4UbKjhLdq6/EuSVR9GMN7KVH3r9gQ83e73hsz1Nd1T3ijd5xv1wcWRYO+D6kCI2w==}
-
   commander@13.0.0:
     resolution: {integrity: sha512-oPYleIY8wmTVzkvQq10AEok6YcTC4sRUBl8F9gVuwchGVUCTbl/vhLTaQqutuuySYOsu8YTgV+OxKc/8Yvx+mQ==}
     engines: {node: '>=18'}
 
-  commander@13.1.0:
-    resolution: {integrity: sha512-/rFeCpNJQbhSZjGVwO9RFV3xPqbnERS8MmIQzCtD/zl6gpJuV/bMLuN92oG3F7d8oDEHHRrujSXNUr8fpjntKw==}
-    engines: {node: '>=18'}
-
   commondir@1.0.1:
     resolution: {integrity: sha512-W9pAhw0ja1Edb5GVdIF1mjZw/ASI0AlShXM83UUGe2DVr5TdAPEA1OA8m/g8zWp9x6On7gqufY+FatDbC3MDQg==}
 
@@ -1031,10 +1007,6 @@ packages:
   emoji-regex@9.2.2:
     resolution: {integrity: sha512-L18DaJsXSUk2+42pv8mLs5jJT2hqFkFE4j21wOmgbUqsZ2hL72NsUU785g9RXgo3s0ZNgVl42TiHp3ZtOv/Vyg==}
 
-  environment@1.1.0:
-    resolution: {integrity: sha512-xUtoPkMggbz0MPyPiIWr1Kp4aeWJjDZ6SMvURhimjdZgsRuDplF5/s9hcgGhyXMhs+6vpnuoiZ2kFiu3FMnS8Q==}
-    engines: {node: '>=18'}
-
   es-module-lexer@1.7.0:
     resolution: {integrity: sha512-jEQoCwk8hyb2AZziIOLhDqpm5+2ww5uIE6lkO/6jcOCusfk6LhMHpXXfBLXTZ7Ydyt0j4VoUQv6uGNYbdW+kBA==}
 
@@ -1099,13 +1071,6 @@ packages:
     resolution: {integrity: sha512-kVscqXk4OCp68SZ0dkgEKVi6/8ij300KBWTJq32P/dYeWTSwK41WyTxalN1eRmA5Z9UU/LX9D7FWSmV9SAYx6g==}
     engines: {node: '>=0.10.0'}
 
-  eventemitter3@5.0.1:
-    resolution: {integrity: sha512-GWkBvjiSZK87ELrYOSESUYeVIc9mvLLf/nXalMOS5dYrgZq9o5OVkbZAVM06CVxYsCwH9BDZFPlQTlPA1j4ahA==}
-
-  execa@8.0.1:
-    resolution: {integrity: sha512-VyhnebXciFV2DESc+p6B+y0LjSm0krU4OgJN44qFAhBY0TJ+1V61tYD2+wHusZ6F9n5K+vl8k0sTy7PEfV4qpg==}
-    engines: {node: '>=16.17'}
-
   expect-type@1.2.2:
     resolution: {integrity: sha512-JhFGDVJ7tmDJItKhYgJCGLOWjuK9vPxiXoUFLwLDc99NlmklilbiQJwoctZtt13+xMw91MCk/REan6MWHqDjyA==}
     engines: {node: '>=12.0.0'}
@@ -1187,14 +1152,6 @@ packages:
     resolution: {integrity: sha512-vpeMIQKxczTD/0s2CdEWHcb0eeJe6TFjxb+J5xgX7hScxqrGuyjmv4c1D4A/gelKfyox0gJJwIHF+fLjeaM8kQ==}
     engines: {node: '>=18'}
 
-  get-east-asian-width@1.3.1:
-    resolution: {integrity: sha512-R1QfovbPsKmosqTnPoRFiJ7CF9MLRgb53ChvMZm+r4p76/+8yKDy17qLL2PKInORy2RkZZekuK0efYgmzTkXyQ==}
-    engines: {node: '>=18'}
-
-  get-stream@8.0.1:
-    resolution: {integrity: sha512-VaUJspBffn/LMCJVoMvSAdmscJyS1auj5Zulnn5UoYcY531UWmdwhRWkcGKnGU93m5HSXP9LP2usOryrBtQowA==}
-    engines: {node: '>=16'}
-
   get-tsconfig@4.10.1:
     resolution: {integrity: sha512-auHyJ4AgMz7vgS8Hp3N6HXSmlMdUyhSUrfBF16w153rxtLIEOE+HGqaBppczZvnHLqQJfiHotCYpNhl0lUROFQ==}
 
@@ -1241,10 +1198,6 @@ packages:
   html-escaper@2.0.2:
     resolution: {integrity: sha512-H2iMtd0I4Mt5eYiapRdIDjp+XzelXQ0tFE4JS7YFwFevXXMmOp9myNrUvCg0D6ws8iqkRPBfKHgbwig1SmlLfg==}
 
-  human-signals@5.0.0:
-    resolution: {integrity: sha512-AXcZb6vzzrFAUE61HnN4mpLqd/cSIwNQjtNWR0euPm6y0iqx3G4gOXaIDdtdDwZmhwe82LA6+zinmW4UBWVePQ==}
-    engines: {node: '>=16.17.0'}
-
   ieee754@1.2.1:
     resolution: {integrity: sha512-dcyqhDvX1C46lXZcVqCpK+FtMRQVdIMN6/Df5js2zouUsqG7I6sFxitIC+7KYK29KdXOLHdu9zL4sFnoVQnqaA==}
 
@@ -1276,14 +1229,6 @@ packages:
     resolution: {integrity: sha512-zymm5+u+sCsSWyD9qNaejV3DFvhCKclKdizYaJUuHA83RLjb7nSuGnddCHGv0hk+KY7BMAlsWeK4Ueg6EV6XQg==}
     engines: {node: '>=8'}
 
-  is-fullwidth-code-point@4.0.0:
-    resolution: {integrity: sha512-O4L094N2/dZ7xqVdrXhh9r1KODPJpFms8B5sGdJLPy664AgvXsreZUyCQQNItZRDlYug4xStLjNp/sz3HvBowQ==}
-    engines: {node: '>=12'}
-
-  is-fullwidth-code-point@5.1.0:
-    resolution: {integrity: sha512-5XHYaSyiqADb4RnZ1Bdad6cPp8Toise4TzEjcOYDHZkTCbKgiUl7WTUCpNWHuxmDt91wnsZBc9xinNzopv3JMQ==}
-    engines: {node: '>=18'}
-
   is-glob@4.0.3:
     resolution: {integrity: sha512-xelSayHH36ZgE7ZWhli7pW34hNbNl8Ojv5KVmkJD4hBdD3th8Tfk9vYasLM+mXWOZhFkgZfxhLSnrwRr4elSSg==}
     engines: {node: '>=0.10.0'}
@@ -1306,10 +1251,6 @@ packages:
   is-reference@1.2.1:
     resolution: {integrity: sha512-U82MsXXiFIrjCK4otLT+o2NA2Cd2g5MLoOVXUZjIOhLurrRxpEXzI8O0KZHr3IjLvlAH1kTPYSuqer5T9ZVBKQ==}
 
-  is-stream@3.0.0:
-    resolution: {integrity: sha512-LnQR4bZ9IADDRSkvpqMGvt/tEJWclzklNgSw48V5EAaAeDd6qGvN8ei6k5p0tvxSR171VmGyHuTiAOfxAbr8kA==}
-    engines: {node: ^12.20.0 || ^14.13.1 || >=16.0.0}
-
   is-unicode-supported@1.3.0:
     resolution: {integrity: sha512-43r2mRvz+8JRIKnWJ+3j8JtjRKZ6GmjzfaE/qiBJnikNnYv/6bagRJ1kUhNk8R5EX/GkobD+r+sfxCPJsiKBLQ==}
     engines: {node: '>=12'}
@@ -1373,19 +1314,6 @@ packages:
     resolution: {integrity: sha512-+bT2uH4E5LGE7h/n3evcS/sQlJXCpIp6ym8OWJ5eV6+67Dsql/LaaT7qJBAt2rzfoa/5QBGBhxDix1dMt2kQKQ==}
     engines: {node: '>= 0.8.0'}
 
-  lilconfig@3.1.3:
-    resolution: {integrity: sha512-/vlFKAoH5Cgt3Ie+JLhRbwOsCQePABiU3tJ1egGvyQ+33R/vcwM2Zl2QR/LzjsBeItPt3oSVXapn+m4nQDvpzw==}
-    engines: {node: '>=14'}
-
-  lint-staged@15.5.2:
-    resolution: {integrity: sha512-YUSOLq9VeRNAo/CTaVmhGDKG+LBtA8KF1X4K5+ykMSwWST1vDxJRB2kv2COgLb1fvpCo+A/y9A0G0znNVmdx4w==}
-    engines: {node: '>=18.12.0'}
-    hasBin: true
-
-  listr2@8.3.3:
-    resolution: {integrity: sha512-LWzX2KsqcB1wqQ4AHgYb4RsDXauQiqhjLk+6hjbaeHG4zpjjVAB6wC/gz6X0l+Du1cN3pUB5ZlrvTbhGSNnUQQ==}
-    engines: {node: '>=18.0.0'}
-
   locate-path@6.0.0:
     resolution: {integrity: sha512-iPZK6eYjbxRu3uB4/WZ3EsEIMJFMqAoopl3R+zuq0UjcAm/MO6KCweDgPfP3elTztoKP3KtnVHxTn2NHBSDVUw==}
     engines: {node: '>=10'}
@@ -1400,10 +1328,6 @@ packages:
     resolution: {integrity: sha512-i24m8rpwhmPIS4zscNzK6MSEhk0DUWa/8iYQWxhffV8jkI4Phvs3F+quL5xvS0gdQR0FyTCMMH33Y78dDTzzIw==}
     engines: {node: '>=18'}
 
-  log-update@6.1.0:
-    resolution: {integrity: sha512-9ie8ItPR6tjY5uYJh8K/Zrv/RMZ5VOlOWvtZdEHYSTFKZfIBPQa9tOAEeAWhd+AnIneLJ22w5fjOYtoutpWq5w==}
-    engines: {node: '>=18'}
-
   loupe@3.2.0:
     resolution: {integrity: sha512-2NCfZcT5VGVNX9mSZIxLRkEAegDGBpuQZBy13desuHeVORmBDyAET4TkJr4SjqQy3A8JDofMN6LpkK8Xcm/dlw==}
 
@@ -1423,9 +1347,6 @@ packages:
     resolution: {integrity: sha512-hXdUTZYIVOt1Ex//jAQi+wTZZpUpwBj/0QsOzqegb3rGMMeJiSEu5xLHnYfBrRV4RH2+OCSOO95Is/7x1WJ4bw==}
     engines: {node: '>=10'}
 
-  merge-stream@2.0.0:
-    resolution: {integrity: sha512-abv/qOcuPfk3URPfDzmZU1LKmuw8kT+0nIHvKrKgFrwifol/doWcdA4ZqsWQ8ENrFKkd67Mfpo/LovbIUsbt3w==}
-
   merge2@1.4.1:
     resolution: {integrity: sha512-8q7VEgMJW4J8tcfVPy8g09NcQwZdbwFEqhe/WZkoIzjn/3TGDwtOCYtXGxA3O8tPzpczCCDgv+P2P5y00ZJOOg==}
     engines: {node: '>= 8'}
@@ -1434,10 +1355,6 @@ packages:
     resolution: {integrity: sha512-PXwfBhYu0hBCPw8Dn0E+WDYb7af3dSLVWKi3HGv84IdF4TyFoC0ysxFd0Goxw7nSv4T/PzEJQxsYsEiFCKo2BA==}
     engines: {node: '>=8.6'}
 
-  mimic-fn@4.0.0:
-    resolution: {integrity: sha512-vqiC06CuhBTUdZH+RYl8sFrL096vA45Ok5ISO6sE/Mr1jRbGH4Csnhi8f3wKVl7x8mO4Au7Ir9D3Oyv1VYMFJw==}
-    engines: {node: '>=12'}
-
   mimic-function@5.0.1:
     resolution: {integrity: sha512-VP79XUPxV2CigYP3jWwAUFSku2aKqBH7uTAapFWCBqutsbmDo96KY5o8uh6U+/YSIn5OxJnXp73beVkpqMIGhA==}
     engines: {node: '>=18'}
@@ -1497,10 +1414,6 @@ packages:
   node-fetch-native@1.6.7:
     resolution: {integrity: sha512-g9yhqoedzIUm0nTnTqAQvueMPVOuIY16bqgAJJC8XOOubYFNwz6IER9qs0Gq2Xd0+CecCKFjtdDTMA4u4xG06Q==}
 
-  npm-run-path@5.3.0:
-    resolution: {integrity: sha512-ppwTtiJZq0O/ai0z7yfudtBpWIoxM8yE6nHi1X47eFR2EWORqfbu6CnPlNsjeN683eT0qG6H/Pyf9fCcvjnnnQ==}
-    engines: {node: ^12.20.0 || ^14.13.1 || >=16.0.0}
-
   nypm@0.5.4:
     resolution: {integrity: sha512-X0SNNrZiGU8/e/zAB7sCTtdxWTMSIO73q+xuKgglm2Yvzwlo8UoC5FNySQFCvl84uPaeADkqHUZUkWy4aH4xOA==}
     engines: {node: ^14.16.0 || >=16.10.0}
@@ -1509,10 +1422,6 @@ packages:
   ohash@1.1.6:
     resolution: {integrity: sha512-TBu7PtV8YkAZn0tSxobKY2n2aAQva936lhRrj6957aDaCf9IEtqsKbgMzXE/F/sjqYOwmrukeORHNLe5glk7Cg==}
 
-  onetime@6.0.0:
-    resolution: {integrity: sha512-1FlR+gjXK7X+AsAHso35MnyN5KqGwJRi/31ft6x0M194ht7S+rWAvd7PHss9xSKMzE0asv1pyIHaJYq+BbacAQ==}
-    engines: {node: '>=12'}
-
   onetime@7.0.0:
     resolution: {integrity: sha512-VXJjc87FScF88uafS3JllDgvAm+c/Slfz06lorj2uAY34rlUu0Nt+v8wreiImcrgAjjIHp1rXpTDlLOGw29WwQ==}
     engines: {node: '>=18'}
@@ -1552,10 +1461,6 @@ packages:
     resolution: {integrity: sha512-ojmeN0qd+y0jszEtoY48r0Peq5dwMEkIlCOu6Q5f41lfkswXuKtYrhgoTpLnyIcHm24Uhqx+5Tqm2InSwLhE6Q==}
     engines: {node: '>=8'}
 
-  path-key@4.0.0:
-    resolution: {integrity: sha512-haREypq7xkM7ErfgIyA0z+Bj4AGKlMSdlQE2jvJo6huWD1EdkKYV+G/T4nq0YEF2vgTT8kqMFKo1uHn950r4SQ==}
-    engines: {node: '>=12'}
-
   path-parse@1.0.7:
     resolution: {integrity: sha512-LDJzPVEEEPR+y48z93A0Ed0yXb8pAByGWo/k5YYdYgpY2/2EsOsksJrq7lOHxryrVOn1ejG6oAp8ahvOIQD8sw==}
 
@@ -1587,11 +1492,6 @@ packages:
     resolution: {integrity: sha512-5gTmgEY/sqK6gFXLIsQNH19lWb4ebPDLA4SdLP7dsWkIXHWlG66oPuVvXSGFPppYZz8ZDZq0dYYrbHfBCVUb1Q==}
     engines: {node: '>=12'}
 
-  pidtree@0.6.0:
-    resolution: {integrity: sha512-eG2dWTVw5bzqGRztnHExczNxt5VGsE6OwTeCG3fdUf9KBsZzO3R5OIIIzWR+iZA0NtZ+RDVdaoE2dK1cn6jH4g==}
-    engines: {node: '>=0.10'}
-    hasBin: true
-
   pkg-types@1.3.1:
     resolution: {integrity: sha512-/Jm5M4RvtBFVkKWRu2BLUTNP8/M2a+UwuAX+ae4770q1qVGtfjG+WTCupoZixokjmHiry8uI+dlY8KXYV5HVVQ==}
 
@@ -1658,9 +1558,6 @@ packages:
     resolution: {integrity: sha512-g6QUff04oZpHs0eG5p83rFLhHeV00ug/Yf9nZM6fLeUrPguBTkTQOdpAWWspMh55TZfVQDPaN3NQJfbVRAxdIw==}
     engines: {iojs: '>=1.0.0', node: '>=0.10.0'}
 
-  rfdc@1.4.1:
-    resolution: {integrity: sha512-q1b3N5QkRUWUl7iyylaaj3kOpIT0N2i9MqIEQXP73GVsN9cw3fdx8X63cEmWhJGi2PPCF23Ijp7ktmd39rawIA==}
-
   rollup-plugin-dts@6.2.1:
     resolution: {integrity: sha512-sR3CxYUl7i2CHa0O7bA45mCrgADyAQ0tVtGSqi3yvH28M+eg1+g5d7kQ9hLvEz5dorK3XVsH5L2jwHLQf72DzA==}
     engines: {node: '>=16'}
@@ -1712,14 +1609,6 @@ packages:
     resolution: {integrity: sha512-FoqMu0NCGBLCcAkS1qA+XJIQTR6/JHfQXl+uGteNCQ76T91DMUjPa9xfmeqMY3z80nLSg9yQmNjK0Px6RWsH/A==}
     engines: {node: '>=18'}
 
-  slice-ansi@5.0.0:
-    resolution: {integrity: sha512-FC+lgizVPfie0kkhqUScwRu1O/lF6NOgJmlCgK+/LYxDCTk8sGelYaHDhFcDN+Sn3Cv+3VSa4Byeo+IMCzpMgQ==}
-    engines: {node: '>=12'}
-
-  slice-ansi@7.1.0:
-    resolution: {integrity: sha512-bSiSngZ/jWeX93BqeIAbImyTbEihizcwNjFoRUIY/T1wWQsfsm2Vw1agPKylXvQTU7iASGdHhyqRlqQzfz+Htg==}
-    engines: {node: '>=18'}
-
   source-map-js@1.2.1:
     resolution: {integrity: sha512-UXWMKhLOwVKb728IUtQPXxfYU+usdybtUrK/8uGE8CQMvrhOpwvzDBwj0QhSL7MQc7vIsISBG8VQ8+IDQxpfQA==}
     engines: {node: '>=0.10.0'}
@@ -1738,10 +1627,6 @@ packages:
     resolution: {integrity: sha512-UhDfHmA92YAlNnCfhmq0VeNL5bDbiZGg7sZ2IvPsXubGkiNa9EC+tUTsjBRsYUAz87btI6/1wf4XoVvQ3uRnmQ==}
     engines: {node: '>=18'}
 
-  string-argv@0.3.2:
-    resolution: {integrity: sha512-aqD2Q0144Z+/RqG52NeHEkZauTAUWJO8c6yTftGJKO3Tja5tUgIfmIl6kExvhtxSDP7fXB6DvzkfMpCd/F3G+Q==}
-    engines: {node: '>=0.6.19'}
-
   string-width@4.2.3:
     resolution: {integrity: sha512-wKyQRQpjJ0sIp62ErSZdGsjMJWsap5oRNihHhu6G7JVO/9jIB6UyevL+tXuOqrng8j/cxKTWyWUwvSTriiZz/g==}
     engines: {node: '>=8'}
@@ -1762,10 +1647,6 @@ packages:
     resolution: {integrity: sha512-iq6eVVI64nQQTRYq2KtEg2d2uU7LElhTJwsH4YzIHZshxlgZms/wIc4VoDQTlG/IvVIrBKG06CrZnp0qv7hkcQ==}
     engines: {node: '>=12'}
 
-  strip-final-newline@3.0.0:
-    resolution: {integrity: sha512-dOESqjYr96iWYylGObzd39EuNTa5VJxyvVAEm5Jnh7KGo75V43Hk1odPQkNDyXNmUR6k+gEiDVXnjB8HJ3crXw==}
-    engines: {node: '>=12'}
-
   strip-json-comments@3.1.1:
     resolution: {integrity: sha512-6fPc+R4ihwqP6N/aIv2f1gMH8lOVtWQHoqC4yK6oSDVVocumAsfCqjkXnqiYMhmMwS/mEHLp7Vehlt3ql6lEig==}
     engines: {node: '>=8'}
@@ -1990,10 +1871,6 @@ packages:
     resolution: {integrity: sha512-si7QWI6zUMq56bESFvagtmzMdGOtoxfR+Sez11Mobfc7tm+VkUckk9bW2UeffTGVUbOksxmSw0AA2gs8g71NCQ==}
     engines: {node: '>=12'}
 
-  wrap-ansi@9.0.0:
-    resolution: {integrity: sha512-G8ura3S+3Z2G+mkgNRq8dqaFZAuxfsxpBB8OCTGRTCtp+l/v9nbFNmCUP1BZMts3G1142MsZfn6eeUKrr4PD1Q==}
-    engines: {node: '>=18'}
-
   y18n@5.0.8:
     resolution: {integrity: sha512-0pfFzegeDWJHJIAmTLRP2DwHjdF5s7jo9tuztdQxAhINCdvS+3nGINqPd00AphqJR/0LhANUS6/+7SCb98YOfA==}
     engines: {node: '>=10'}
@@ -2001,11 +1878,6 @@ packages:
   yallist@4.0.0:
     resolution: {integrity: sha512-3wdGidZyq5PB084XLES5TpOSRA3wjXAlIWMhum2kRcv/41Sn2emQ0dycQW4uZXLejwKvg6EsvbdlVL+FYEct7A==}
 
-  yaml@2.8.1:
-    resolution: {integrity: sha512-lcYcMxX2PO9XMGvAJkJ3OsNMw+/7FKes7/hgerGUYWIoWu5j/+YQqcZr5JnPZWzOsEBgMbSbiSTn/dv/69Mkpw==}
-    engines: {node: '>= 14.6'}
-    hasBin: true
-
   yargs-parser@21.1.1:
     resolution: {integrity: sha512-tVpsJW7DdjecAiFpbIB1e3qxIQsE6NoPc5/eTdrbbIC4h0LVsWhnoa3g+m2HclBIujHzsxZ4VJVA+GUuc2/LBw==}
     engines: {node: '>=12'}
@@ -2679,10 +2551,6 @@ snapshots:
       json-schema-traverse: 1.0.0
       require-from-string: 2.0.2
 
-  ansi-escapes@7.0.0:
-    dependencies:
-      environment: 1.1.0
-
   ansi-regex@5.0.1: {}
 
   ansi-regex@6.1.0: {}
@@ -2797,11 +2665,6 @@ snapshots:
 
   cli-spinners@2.9.2: {}
 
-  cli-truncate@4.0.0:
-    dependencies:
-      slice-ansi: 5.0.0
-      string-width: 7.2.0
-
   cliui@8.0.1:
     dependencies:
       string-width: 4.2.3
@@ -2814,12 +2677,8 @@ snapshots:
 
   color-name@1.1.4: {}
 
-  colorette@2.0.20: {}
-
   commander@13.0.0: {}
 
-  commander@13.1.0: {}
-
   commondir@1.0.1: {}
 
   concat-map@0.0.1: {}
@@ -2858,8 +2717,6 @@ snapshots:
 
   emoji-regex@9.2.2: {}
 
-  environment@1.1.0: {}
-
   es-module-lexer@1.7.0: {}
 
   esbuild@0.21.5:
@@ -2967,20 +2824,6 @@ snapshots:
 
   esutils@2.0.3: {}
 
-  eventemitter3@5.0.1: {}
-
-  execa@8.0.1:
-    dependencies:
-      cross-spawn: 7.0.6
-      get-stream: 8.0.1
-      human-signals: 5.0.0
-      is-stream: 3.0.0
-      merge-stream: 2.0.0
-      npm-run-path: 5.3.0
-      onetime: 6.0.0
-      signal-exit: 4.1.0
-      strip-final-newline: 3.0.0
-
   expect-type@1.2.2: {}
 
   fast-deep-equal@3.1.3: {}
@@ -3056,10 +2899,6 @@ snapshots:
 
   get-east-asian-width@1.3.0: {}
 
-  get-east-asian-width@1.3.1: {}
-
-  get-stream@8.0.1: {}
-
   get-tsconfig@4.10.1:
     dependencies:
       resolve-pkg-maps: 1.0.0
@@ -3114,8 +2953,6 @@ snapshots:
 
   html-escaper@2.0.2: {}
 
-  human-signals@5.0.0: {}
-
   ieee754@1.2.1: {}
 
   ignore@5.3.2: {}
@@ -3137,12 +2974,6 @@ snapshots:
 
   is-fullwidth-code-point@3.0.0: {}
 
-  is-fullwidth-code-point@4.0.0: {}
-
-  is-fullwidth-code-point@5.1.0:
-    dependencies:
-      get-east-asian-width: 1.3.1
-
   is-glob@4.0.3:
     dependencies:
       is-extglob: 2.1.1
@@ -3159,8 +2990,6 @@ snapshots:
     dependencies:
       '@types/estree': 1.0.8
 
-  is-stream@3.0.0: {}
-
   is-unicode-supported@1.3.0: {}
 
   is-unicode-supported@2.1.0: {}
@@ -3224,32 +3053,6 @@ snapshots:
       prelude-ls: 1.2.1
       type-check: 0.4.0
 
-  lilconfig@3.1.3: {}
-
-  lint-staged@15.5.2:
-    dependencies:
-      chalk: 5.5.0
-      commander: 13.1.0
-      debug: 4.4.1
-      execa: 8.0.1
-      lilconfig: 3.1.3
-      listr2: 8.3.3
-      micromatch: 4.0.8
-      pidtree: 0.6.0
-      string-argv: 0.3.2
-      yaml: 2.8.1
-    transitivePeerDependencies:
-      - supports-color
-
-  listr2@8.3.3:
-    dependencies:
-      cli-truncate: 4.0.0
-      colorette: 2.0.20
-      eventemitter3: 5.0.1
-      log-update: 6.1.0
-      rfdc: 1.4.1
-      wrap-ansi: 9.0.0
-
   locate-path@6.0.0:
     dependencies:
       p-locate: 5.0.0
@@ -3263,14 +3066,6 @@ snapshots:
       chalk: 5.5.0
       is-unicode-supported: 1.3.0
 
-  log-update@6.1.0:
-    dependencies:
-      ansi-escapes: 7.0.0
-      cli-cursor: 5.0.0
-      slice-ansi: 7.1.0
-      strip-ansi: 7.1.0
-      wrap-ansi: 9.0.0
-
   loupe@3.2.0: {}
 
   lru-cache@10.4.3: {}
@@ -3291,8 +3086,6 @@ snapshots:
     dependencies:
       semver: 7.7.2
 
-  merge-stream@2.0.0: {}
-
   merge2@1.4.1: {}
 
   micromatch@4.0.8:
@@ -3300,8 +3093,6 @@ snapshots:
       braces: 3.0.3
       picomatch: 2.3.1
 
-  mimic-fn@4.0.0: {}
-
   mimic-function@5.0.1: {}
 
   minimatch@3.1.2:
@@ -3348,10 +3139,6 @@ snapshots:
 
   node-fetch-native@1.6.7: {}
 
-  npm-run-path@5.3.0:
-    dependencies:
-      path-key: 4.0.0
-
   nypm@0.5.4:
     dependencies:
       citty: 0.1.6
@@ -3363,10 +3150,6 @@ snapshots:
 
   ohash@1.1.6: {}
 
-  onetime@6.0.0:
-    dependencies:
-      mimic-fn: 4.0.0
-
   onetime@7.0.0:
     dependencies:
       mimic-function: 5.0.1
@@ -3416,8 +3199,6 @@ snapshots:
 
   path-key@3.1.1: {}
 
-  path-key@4.0.0: {}
-
   path-parse@1.0.7: {}
 
   path-scurry@1.11.1:
@@ -3439,8 +3220,6 @@ snapshots:
 
   picomatch@4.0.3: {}
 
-  pidtree@0.6.0: {}
-
   pkg-types@1.3.1:
     dependencies:
       confbox: 0.1.8
@@ -3493,8 +3272,6 @@ snapshots:
 
   reusify@1.1.0: {}
 
-  rfdc@1.4.1: {}
-
   rollup-plugin-dts@6.2.1(rollup@4.46.2)(typescript@5.9.2):
     dependencies:
       magic-string: 0.30.17
@@ -3565,16 +3342,6 @@ snapshots:
       mrmime: 2.0.1
       totalist: 3.0.1
 
-  slice-ansi@5.0.0:
-    dependencies:
-      ansi-styles: 6.2.1
-      is-fullwidth-code-point: 4.0.0
-
-  slice-ansi@7.1.0:
-    dependencies:
-      ansi-styles: 6.2.1
-      is-fullwidth-code-point: 5.1.0
-
   source-map-js@1.2.1: {}
 
   source-map@0.6.1: {}
@@ -3585,8 +3352,6 @@ snapshots:
 
   stdin-discarder@0.2.2: {}
 
-  string-argv@0.3.2: {}
-
   string-width@4.2.3:
     dependencies:
       emoji-regex: 8.0.0
@@ -3613,8 +3378,6 @@ snapshots:
     dependencies:
       ansi-regex: 6.1.0
 
-  strip-final-newline@3.0.0: {}
-
   strip-json-comments@3.1.1: {}
 
   strtok3@10.3.4:
@@ -3821,18 +3584,10 @@ snapshots:
       string-width: 5.1.2
       strip-ansi: 7.1.0
 
-  wrap-ansi@9.0.0:
-    dependencies:
-      ansi-styles: 6.2.1
-      string-width: 7.2.0
-      strip-ansi: 7.1.0
-
   y18n@5.0.8: {}
 
   yallist@4.0.0: {}
 
-  yaml@2.8.1: {}
-
   yargs-parser@21.1.1: {}
 
   yargs@17.7.2:

py/README.md

@@ -9,6 +9,7 @@ This repository contains the code for hand-written SDKs and clients for interact
 This includes:
 
 - [LlamaParse](../parse.md) - A GenAI-native document parser that can parse complex document data for any downstream LLM use case (Agents, RAG, data processing, etc.).
+- [LlamaReport (beta/invite-only)](../report.md) - A prebuilt agentic report builder that can be used to build reports from a variety of data sources.
 - [LlamaExtract](../extract.md) - A prebuilt agentic data extractor that can be used to transform data into a structured JSON representation.
 - [LlamaCloud Index](../index.md) - A widely customizable and fully automated document ingestion pipeline that also serves retrieval purposes.
 
@@ -27,12 +28,14 @@ Then, you can use the services in your code:
 ```python
 from llama_cloud_services import (
     LlamaParse,
+    LlamaReport,
     LlamaExtract,
     LlamaCloudIndex,
 )
-from llama_cloud_services import LlamaParse, LlamaExtract
+from llama_cloud_services import LlamaParse, LlamaReport, LlamaExtract
 
 parser = LlamaParse(api_key="YOUR_API_KEY")
+report = LlamaReport(api_key="YOUR_API_KEY")
 extract = LlamaExtract(api_key="YOUR_API_KEY")
 index = LlamaCloudIndex(
     "my_first_index", project_name="default", api_key="YOUR_API_KEY"
@@ -42,6 +45,7 @@ index = LlamaCloudIndex(
 See the quickstart guides for each service for more information:
 
 - [LlamaParse](../parse.md)
+- [LlamaReport (beta/invite-only)](../report.md)
 - [LlamaExtract](../extract.md)
 - [LlamaCloud Index](../index.md)
 
@@ -54,11 +58,13 @@ You can also create your API key in the EU region [here](https://cloud.eu.llamai
 ```python
 from llama_cloud_services import (
     LlamaParse,
+    LlamaReport,
     LlamaExtract,
     EU_BASE_URL,
 )
 
 parser = LlamaParse(api_key="YOUR_API_KEY", base_url=EU_BASE_URL)
+report = LlamaReport(api_key="YOUR_API_KEY", base_url=EU_BASE_URL)
 extract = LlamaExtract(api_key="YOUR_API_KEY", base_url=EU_BASE_URL)
 index = LlamaCloudIndex(
     "my_first_index",

py/llama_cloud_services/__init__.py

@@ -1,5 +1,6 @@
 from llama_cloud_services.parse import LlamaParse
-from llama_cloud_services.extract import LlamaExtract, ExtractionAgent, SourceText
+from llama_cloud_services.report import ReportClient, LlamaReport
+from llama_cloud_services.extract import LlamaExtract, ExtractionAgent
 from llama_cloud_services.constants import EU_BASE_URL
 from llama_cloud_services.index import (
     LlamaCloudCompositeRetriever,
@@ -9,9 +10,10 @@ from llama_cloud_services.index import (
 
 __all__ = [
     "LlamaParse",
+    "ReportClient",
+    "LlamaReport",
     "LlamaExtract",
     "ExtractionAgent",
-    "SourceText",
     "EU_BASE_URL",
     "LlamaCloudIndex",
     "LlamaCloudRetriever",

py/llama_cloud_services/beta/classifier/client.py

@@ -17,9 +17,6 @@ from llama_cloud_services.files.client import FileClient
 from llama_cloud_services.constants import POLLING_TIMEOUT_SECONDS
 from llama_cloud_services.utils import is_terminal_status, augment_async_errors
 from llama_index.core.async_utils import DEFAULT_NUM_WORKERS, run_jobs
-from llama_cloud_services.beta.classifier.types import (
-    ClassifyJobResultsWithFiles,
-)
 
 
 class ClassificationOutput(BaseModel):
@@ -55,24 +52,6 @@ class ClassifyClient:
         self.file_client = FileClient(client, project_id, organization_id)
         self.polling_timeout = polling_timeout
 
-    @classmethod
-    def from_api_key(
-        cls,
-        api_key: str,
-        project_id: Optional[str] = None,
-        organization_id: Optional[str] = None,
-        base_url: Optional[str] = None,
-    ) -> "ClassifyClient":
-        """
-        Create a classify client from an API key.
-        """
-        client = AsyncLlamaCloud(token=api_key, base_url=base_url)
-        return cls(
-            client,
-            project_id,
-            organization_id,
-        )
-
     async def acreate_classify_job(
         self,
         rules: list[ClassifierRule],
@@ -173,12 +152,11 @@ class ClassifyClient:
         file_input_path: str,
         parsing_configuration: Optional[ClassifyParsingConfiguration] = None,
         raise_on_error: bool = True,
-    ) -> ClassifyJobResultsWithFiles:
+    ) -> ClassifyJobResults:
         file = await self.file_client.upload_file(file_input_path)
-        results = await self.aclassify_file_ids(
+        return await self.aclassify_file_ids(
             rules, [file.id], parsing_configuration, raise_on_error
         )
-        return ClassifyJobResultsWithFiles.from_classify_job_results(results, [file])
 
     def classify_file_path(
         self,
@@ -186,7 +164,7 @@ class ClassifyClient:
         file_input_path: str,
         parsing_configuration: Optional[ClassifyParsingConfiguration] = None,
         raise_on_error: bool = True,
-    ) -> ClassifyJobResultsWithFiles:
+    ) -> ClassifyJobResults:
         with augment_async_errors():
             return asyncio.run(
                 self.aclassify_file_path(
@@ -202,7 +180,7 @@ class ClassifyClient:
         raise_on_error: bool = True,
         workers: int = DEFAULT_NUM_WORKERS,
         show_progress: bool = False,
-    ) -> ClassifyJobResultsWithFiles:
+    ) -> ClassifyJobResults:
         coroutines = [self.file_client.upload_file(path) for path in file_input_paths]
         files: list[File] = await run_jobs(
             coroutines,
@@ -210,10 +188,9 @@ class ClassifyClient:
             workers=workers,
             desc="Uploading files for classification",
         )
-        results = await self.aclassify_file_ids(
+        return await self.aclassify_file_ids(
             rules, [file.id for file in files], parsing_configuration, raise_on_error
         )
-        return ClassifyJobResultsWithFiles.from_classify_job_results(results, files)
 
     def classify_file_paths(
         self,
@@ -221,7 +198,7 @@ class ClassifyClient:
         file_input_paths: list[str],
         parsing_configuration: Optional[ClassifyParsingConfiguration] = None,
         raise_on_error: bool = True,
-    ) -> ClassifyJobResultsWithFiles:
+    ) -> ClassifyJobResults:
         with augment_async_errors():
             return asyncio.run(
                 self.aclassify_file_paths(

py/llama_cloud_services/beta/classifier/types.py

@@ -1,59 +0,0 @@
-from llama_cloud.types.classify_job_results import ClassifyJobResults
-from llama_cloud.types.file_classification import FileClassification
-from llama_cloud.types.file import File
-
-
-class FileClassificationWithFile(FileClassification):
-    """
-    File classification with file object.
-    """
-
-    file: File
-
-    @classmethod
-    def from_file_classification(
-        cls, file_classification: FileClassification, file: File
-    ) -> "FileClassificationWithFile":
-        if file_classification.file_id != file.id:
-            raise ValueError(
-                f"File classification ID {file_classification.id} does not match file ID {file.id}"
-            )
-        ctor_args = {
-            **file_classification.dict(),
-            "file": file,
-        }
-        return cls(**ctor_args)
-
-
-class ClassifyJobResultsWithFiles(ClassifyJobResults):
-    """
-    Classify job results with file objects.
-    """
-
-    items: list[FileClassificationWithFile]
-
-    @classmethod
-    def from_classify_job_results(
-        cls, classify_job_results: ClassifyJobResults, files: list[File]
-    ) -> "ClassifyJobResultsWithFiles":
-        if len(classify_job_results.items) != len(files):
-            raise ValueError(
-                f"Number of classify job results {len(classify_job_results.items)} does not match number of files {len(files)}"
-            )
-        # create mapping of file classification result to file object
-        file_id_to_file: dict[str, File] = {file.id: file for file in files}
-        file_classification_to_file: list[tuple[FileClassification, File]] = []
-        for item in classify_job_results.items:
-            if item.file_id not in file_id_to_file:
-                raise ValueError(
-                    f"File classification result {item.id} has file ID {item.file_id} that does not match any provided file ID"
-                )
-            file_classification_to_file.append((item, file_id_to_file[item.file_id]))
-
-        # create a list of file classification with file objects
-        ctor_args = classify_job_results.dict()
-        ctor_args["items"] = [
-            FileClassificationWithFile.from_file_classification(item, file)
-            for item, file in file_classification_to_file
-        ]
-        return cls(**ctor_args)

py/llama_cloud_services/parse/base.py

@@ -1015,11 +1015,20 @@ class LlamaParse(BasePydanticReader):
         try:
             url = build_url(JOB_UPLOAD_ROUTE, self.organization_id, self.project_id)
             resp = await make_api_request(self.aclient, "POST", url, timeout=self.max_timeout, files=files, data=data)  # type: ignore
-            resp.raise_for_status()  # this raises if status is not 2xx
+            # Note: make_api_request already calls raise_for_status(), so no need to call it again
             return resp.json()["id"]
-        except httpx.HTTPStatusError as err:  # this catches it
+        except httpx.HTTPStatusError as err:  # this catches HTTP status errors
             msg = f"Failed to parse the file: {err.response.text}"
             raise Exception(msg) from err  # this preserves the exception context
+        except Exception as err:  # this catches other exceptions like RetryError, ValueError, etc.
+            # Try to extract meaningful error message from the exception chain
+            if hasattr(err, '__cause__') and isinstance(err.__cause__, httpx.HTTPStatusError):
+                # If the exception was caused by an HTTPStatusError, extract the response text
+                msg = f"Failed to parse the file: {err.__cause__.response.text}"
+            else:
+                # For other exceptions, use the string representation
+                msg = f"Failed to parse the file: {str(err)}"
+            raise Exception(msg) from err
         finally:
             if file_handle is not None:
                 file_handle.close()
@@ -1578,7 +1587,7 @@ class LlamaParse(BasePydanticReader):
                             resp = await make_api_request(
                                 client, "GET", asset_url, timeout=self.max_timeout
                             )
-                            resp.raise_for_status()
+                            # Note: make_api_request already calls raise_for_status()
                             f.write(resp.content)
                         assets.append(asset)
             return assets
@@ -1676,7 +1685,7 @@ class LlamaParse(BasePydanticReader):
                     res = await make_api_request(
                         client, "GET", xlsx_url, timeout=self.max_timeout
                     )
-                    res.raise_for_status()
+                    # Note: make_api_request already calls raise_for_status()
                     f.write(res.content)
                 xlsx_list.append(xlsx)
             return xlsx_list

py/llama_cloud_services/parse/types.py

@@ -159,9 +159,6 @@ class Page(BaseModel):
     durationInSeconds: Optional[float] = Field(
         default=None, description="The duration of the audio transcript in seconds."
     )
-    slideSpeakerNotes: Optional[str] = Field(
-        default=None, description="The speaker notes for the slide."
-    )
 
 
 class JobResult(BaseModel):

py/llama_cloud_services/parse/utils.py

@@ -11,6 +11,7 @@ from tenacity import (
     wait_exponential,
     retry_if_exception,
     before_sleep_log,
+    RetryError,
 )
 from typing import Any, Iterable, Iterator, Optional, List, cast
 
@@ -300,7 +301,17 @@ async def make_api_request(
         response.raise_for_status()
         return response
 
-    return await _make_request(url, **httpx_kwargs)
+    try:
+        return await _make_request(url, **httpx_kwargs)
+    except RetryError as retry_err:
+        # Extract the last exception from the retry error to preserve the original error details
+        if retry_err.last_attempt and retry_err.last_attempt.exception():
+            last_exception = retry_err.last_attempt.exception()
+            # Re-raise the original exception to preserve error details like response.text
+            raise last_exception from retry_err
+        else:
+            # Fallback if we can't extract the original exception
+            raise retry_err
 
 
 def expand_target_pages(target_pages: str) -> Iterator[int]:

py/llama_cloud_services/report/__init__.py

@@ -0,0 +1,4 @@
+from llama_cloud_services.report.report import ReportClient
+from llama_cloud_services.report.base import LlamaReport
+
+__all__ = ["ReportClient", "LlamaReport"]

py/llama_cloud_services/report/base.py

@@ -0,0 +1,269 @@
+import asyncio
+import httpx
+import os
+import io
+from concurrent.futures import ThreadPoolExecutor
+from typing import Optional, List, Union, Any, Coroutine, TypeVar
+from urllib.parse import urljoin
+
+from llama_cloud.types import ReportMetadata
+from llama_cloud_services.report.report import ReportClient
+
+T = TypeVar("T")
+
+
+class LlamaReport:
+    """Client for managing reports and general report operations."""
+
+    def __init__(
+        self,
+        api_key: Optional[str] = None,
+        project_id: Optional[str] = None,
+        organization_id: Optional[str] = None,
+        base_url: Optional[str] = None,
+        timeout: Optional[int] = None,
+        async_httpx_client: Optional[httpx.AsyncClient] = None,
+    ):
+        self.api_key = api_key or os.getenv("LLAMA_CLOUD_API_KEY", None)
+        if not self.api_key:
+            raise ValueError("No API key provided.")
+
+        self.base_url = base_url or os.getenv(
+            "LLAMA_CLOUD_BASE_URL", "https://api.cloud.llamaindex.ai"
+        )
+        self.timeout = timeout or 60
+
+        # Initialize HTTP clients
+        self._aclient = async_httpx_client or httpx.AsyncClient(timeout=self.timeout)
+
+        # Set auth headers
+        self.headers = {
+            "Authorization": f"Bearer {self.api_key}",
+        }
+
+        self.organization_id = organization_id
+        self.project_id = project_id
+        self._client_params = {
+            "timeout": self._aclient.timeout,
+            "headers": self._aclient.headers,
+            "base_url": self._aclient.base_url,
+            "auth": self._aclient.auth,
+            "event_hooks": self._aclient.event_hooks,
+            "cookies": self._aclient.cookies,
+            "max_redirects": self._aclient.max_redirects,
+            "params": self._aclient.params,
+            "trust_env": self._aclient.trust_env,
+        }
+        self._thread_pool = ThreadPoolExecutor(
+            max_workers=min(10, (os.cpu_count() or 1) + 4)
+        )
+
+    @property
+    def aclient(self) -> httpx.AsyncClient:
+        if self._aclient is None:
+            self._aclient = httpx.AsyncClient(**self._client_params)
+        return self._aclient
+
+    def _run_sync(self, coro: Coroutine[Any, Any, T]) -> T:
+        """Run coroutine in a separate thread to avoid event loop issues"""
+
+        # force a new client for this thread/event loop
+        original_client = self._aclient
+        self._aclient = None
+
+        def run_coro() -> T:
+            async def wrapped_coro() -> T:
+                return await coro
+
+            return asyncio.run(wrapped_coro())
+
+        result = self._thread_pool.submit(run_coro).result()
+
+        # restore the original client
+        self._aclient = original_client
+
+        return result
+
+    async def _get_default_project(self) -> str:
+        response = await self.aclient.get(
+            urljoin(str(self.base_url), "/api/v1/projects"), headers=self.headers
+        )
+        response.raise_for_status()
+        projects = response.json()
+        default_project = [p for p in projects if p.get("is_default")]
+        return default_project[0]["id"]
+
+    async def _build_url(
+        self, endpoint: str, extra_params: Optional[List[str]] = None
+    ) -> str:
+        """Helper method to build URLs with common query parameters."""
+        url = urljoin(str(self.base_url), endpoint)
+
+        if not self.project_id:
+            self.project_id = await self._get_default_project()
+
+        query_params = []
+        if self.organization_id:
+            query_params.append(f"organization_id={self.organization_id}")
+        if self.project_id:
+            query_params.append(f"project_id={self.project_id}")
+        if extra_params:
+            query_params.extend([p for p in extra_params if p is not None])
+
+        if query_params:
+            url += "?" + "&".join(query_params)
+
+        return url
+
+    async def acreate_report(
+        self,
+        name: str,
+        template_instructions: Optional[str] = None,
+        template_text: Optional[str] = None,
+        template_file: Optional[Union[str, tuple[str, bytes]]] = None,
+        input_files: Optional[List[Union[str, tuple[str, bytes]]]] = None,
+        existing_retriever_id: Optional[str] = None,
+    ) -> ReportClient:
+        """Create a new report asynchronously."""
+        url = await self._build_url("/api/v1/reports/")
+        open_files: List[io.BufferedReader] = []
+
+        data = {"name": name}
+        if template_instructions:
+            data["template_instructions"] = template_instructions
+        if template_text:
+            data["template_text"] = template_text
+        if existing_retriever_id:
+            data["existing_retriever_id"] = str(existing_retriever_id)
+
+        files: List[tuple[str, io.BufferedReader | bytes]] = []
+        if template_file:
+            if isinstance(template_file, str):
+                open_files.append(open(template_file, "rb"))
+                files.append(("template_file", open_files[-1]))
+            else:
+                files.append(("template_file", template_file[1]))
+
+        if input_files:
+            for f in input_files:
+                if isinstance(f, str):
+                    open_files.append(open(f, "rb"))
+                    files.append(("files", open_files[-1]))
+                else:
+                    files.append(("files", f[1]))
+
+        response = await self.aclient.post(
+            url, headers=self.headers, data=data, files=files
+        )
+        try:
+            response.raise_for_status()
+            report_id = response.json()["id"]
+            return ReportClient(report_id, name, self)
+        except httpx.HTTPStatusError as e:
+            raise ValueError(
+                f"Failed to create report: {e.response.text}\nError Code: {e.response.status_code}"
+            )
+        finally:
+            for open_file in open_files:
+                open_file.close()
+
+    def create_report(
+        self,
+        name: str,
+        template_instructions: Optional[str] = None,
+        template_text: Optional[str] = None,
+        template_file: Optional[Union[str, tuple[str, bytes]]] = None,
+        input_files: Optional[List[Union[str, tuple[str, bytes]]]] = None,
+        existing_retriever_id: Optional[str] = None,
+    ) -> ReportClient:
+        """Create a new report."""
+        return self._run_sync(
+            self.acreate_report(
+                name=name,
+                template_instructions=template_instructions,
+                template_text=template_text,
+                template_file=template_file,
+                input_files=input_files,
+                existing_retriever_id=existing_retriever_id,
+            )
+        )
+
+    async def alist_reports(
+        self, state: Optional[str] = None, limit: int = 100, offset: int = 0
+    ) -> List[ReportClient]:
+        """List all reports asynchronously."""
+        params = []
+        if state:
+            params.append(f"state={state}")
+        if limit:
+            params.append(f"limit={limit}")
+        if offset:
+            params.append(f"offset={offset}")
+
+        url = await self._build_url(
+            "/api/v1/reports/list",
+            extra_params=params,
+        )
+
+        response = await self.aclient.get(url, headers=self.headers)
+        response.raise_for_status()
+        data = response.json()
+
+        return [
+            ReportClient(r["report_id"], r["name"], self)
+            for r in data["report_responses"]
+        ]
+
+    def list_reports(
+        self, state: Optional[str] = None, limit: int = 100, offset: int = 0
+    ) -> List[ReportClient]:
+        """Synchronous wrapper for listing reports."""
+        return self._run_sync(self.alist_reports(state, limit, offset))
+
+    async def aget_report(self, report_id: str) -> ReportClient:
+        """Get a Report instance for working with a specific report."""
+        url = await self._build_url(f"/api/v1/reports/{report_id}")
+
+        response = await self.aclient.get(url, headers=self.headers)
+        response.raise_for_status()
+        data = response.json()
+
+        return ReportClient(data["report_id"], data["name"], self)
+
+    def get_report(self, report_id: str) -> ReportClient:
+        """Synchronous wrapper for getting a report."""
+        return self._run_sync(self.aget_report(report_id))
+
+    async def aget_report_metadata(self, report_id: str) -> ReportMetadata:
+        """Get metadata for a specific report asynchronously.
+
+        Returns:
+            dict containing:
+            - id: Report ID
+            - name: Report name
+            - state: Current report state
+            - report_metadata: Additional metadata
+            - template_file: Name of template file if used
+            - template_instructions: Template instructions if provided
+            - input_files: List of input file names
+        """
+        url = await self._build_url(f"/api/v1/reports/{report_id}/metadata")
+
+        response = await self.aclient.get(url, headers=self.headers)
+        response.raise_for_status()
+        return ReportMetadata(**response.json())
+
+    def get_report_metadata(self, report_id: str) -> ReportMetadata:
+        """Synchronous wrapper for getting report metadata."""
+        return self._run_sync(self.aget_report_metadata(report_id))
+
+    async def adelete_report(self, report_id: str) -> None:
+        """Delete a specific report asynchronously."""
+        url = await self._build_url(f"/api/v1/reports/{report_id}")
+
+        response = await self.aclient.delete(url, headers=self.headers)
+        response.raise_for_status()
+
+    def delete_report(self, report_id: str) -> None:
+        """Synchronous wrapper for deleting a report."""
+        return self._run_sync(self.adelete_report(report_id))

py/llama_cloud_services/report/report.py

@@ -0,0 +1,527 @@
+import asyncio
+import httpx
+import time
+from typing import Optional, List, Literal, Union, TYPE_CHECKING
+from dataclasses import dataclass
+from datetime import datetime
+from enum import Enum
+
+from llama_cloud.types import (
+    ReportEventItemEventData_Progress,
+    ReportMetadata,
+    EditSuggestion,
+    ReportResponse,
+    ReportPlan,
+    ReportBlock,
+    ReportPlanBlock,
+    Report,
+)
+
+if TYPE_CHECKING:
+    from llama_cloud_services.report.base import LlamaReport
+
+
+class MessageRole(str, Enum):
+    USER = "user"
+    ASSISTANT = "assistant"
+
+
+@dataclass
+class Message:
+    role: MessageRole
+    content: str
+    timestamp: datetime
+
+
+@dataclass
+class EditAction:
+    block_idx: int
+    old_content: str
+    new_content: Optional[str]
+    action: Literal["approved", "rejected"]
+    timestamp: datetime
+
+
+DEFAULT_POLL_INTERVAL = 5
+DEFAULT_TIMEOUT = 600
+
+
+class ReportClient:
+    """Client for operations on a specific report."""
+
+    def __init__(self, report_id: str, name: str, parent_client: "LlamaReport"):
+        self.report_id = report_id
+        self.name = name
+        self._client = parent_client
+        self._headers = parent_client.headers
+        self._run_sync = parent_client._run_sync
+        self._build_url = parent_client._build_url
+        self.chat_history: List[Message] = []
+        self.edit_history: List[EditAction] = []
+
+    @property
+    def aclient(self) -> httpx.AsyncClient:
+        return self._client.aclient
+
+    def __str__(self) -> str:
+        return f"Report(id={self.report_id}, name={self.name})"
+
+    def __repr__(self) -> str:
+        return f"Report(id={self.report_id}, name={self.name})"
+
+    def _get_block_content(self, block: Union[ReportBlock, ReportPlanBlock]) -> str:
+        if isinstance(block, ReportBlock):
+            return block.template
+        elif isinstance(block, ReportPlanBlock):
+            return block.block.template
+        else:
+            raise ValueError(f"Invalid block type: {type(block)}")
+
+    def _get_block_idx(self, block: Union[ReportBlock, ReportPlanBlock]) -> int:
+        if isinstance(block, ReportBlock):
+            return block.idx
+        elif isinstance(block, ReportPlanBlock):
+            return block.block.idx
+        else:
+            raise ValueError(f"Invalid block type: {type(block)}")
+
+    async def aget(self, version: Optional[int] = None) -> ReportResponse:
+        """Get this report's details asynchronously."""
+        extra_params = []
+        if version is not None:
+            extra_params.append(f"version={version}")
+
+        url = await self._build_url(f"/api/v1/reports/{self.report_id}", extra_params)
+
+        response = await self.aclient.get(url, headers=self._headers)
+        response.raise_for_status()
+        return ReportResponse(**response.json())
+
+    def get(self, version: Optional[int] = None) -> ReportResponse:
+        """Synchronous wrapper for getting this report's details."""
+        return self._run_sync(self.aget(version))
+
+    async def aupdate_report(self, updated_report: Report) -> ReportResponse:
+        """Update this report's content asynchronously."""
+        url = await self._build_url(f"/api/v1/reports/{self.report_id}")
+        response = await self.aclient.patch(
+            url, headers=self._headers, json={"content": updated_report.dict()}
+        )
+        response.raise_for_status()
+        return ReportResponse(**response.json())
+
+    def update_report(self, updated_report: Report) -> ReportResponse:
+        """Synchronous wrapper for updating this report's content."""
+        return self._run_sync(self.aupdate_report(updated_report))
+
+    async def aupdate_plan(
+        self,
+        action: Literal["approve", "reject", "edit"],
+        updated_plan: Optional[ReportPlan] = None,
+    ) -> ReportResponse:
+        """Update this report's plan asynchronously."""
+        if action == "edit" and not updated_plan:
+            raise ValueError("updated_plan is required when action is 'edit'")
+
+        url = await self._build_url(
+            f"/api/v1/reports/{self.report_id}/plan", [f"action={action}"]
+        )
+
+        data = None
+        if updated_plan is not None:
+            plan_dict = updated_plan.dict()
+            plan_dict.pop("generated_at", None)
+            data = plan_dict
+
+        if updated_plan is None and action == "edit":
+            raise ValueError("updated_plan is required when action is 'edit'")
+
+        response = await self.aclient.patch(url, headers=self._headers, json=data)
+        response.raise_for_status()
+        return ReportResponse(**response.json())
+
+    def update_plan(
+        self,
+        action: Literal["approve", "reject", "edit"],
+        updated_plan: Optional[ReportPlan] = None,
+    ) -> ReportResponse:
+        """Synchronous wrapper for updating this report's plan."""
+        return self._run_sync(self.aupdate_plan(action, updated_plan))
+
+    async def asuggest_edits(
+        self,
+        user_query: str,
+        auto_history: bool = True,
+        chat_history: Optional[List[dict]] = None,
+    ) -> List[EditSuggestion]:
+        """Get AI suggestions for edits to this report asynchronously.
+
+        Args:
+            user_query: The user's request/question about what to edit
+            auto_history: Whether to automatically add the user's message to the chat history
+            chat_history:
+                A list of chat messages to include in the chat history.
+                The format being a list of dictionaries with "role" and "content" keys.
+        """
+        # Add user message to history
+        self.chat_history.append(
+            Message(role=MessageRole.USER, content=user_query, timestamp=datetime.now())
+        )
+
+        # Format chat history with edit summaries
+        chat_history_dicts = []
+        for msg in self.chat_history[:-1]:  # Exclude current message
+            content = msg.content
+            if msg.role == MessageRole.USER:
+                # Add edit summary for user messages
+                edit_summary = self._get_edit_summary_after_message(msg.timestamp)
+                if edit_summary:
+                    content = f"{content}\n\nActions taken:\n{edit_summary}"
+
+            chat_history_dicts.append({"role": msg.role.value, "content": content})
+
+        # decide whether to include chat history or not
+        if chat_history:
+            chat_history_dicts = chat_history
+        elif auto_history:
+            chat_history_dicts = chat_history_dicts
+        else:
+            chat_history_dicts = []
+
+        # Make the API call
+        url = await self._build_url(f"/api/v1/reports/{self.report_id}/suggest_edits")
+        data = {"user_query": user_query, "chat_history": chat_history_dicts}
+
+        response = await self.aclient.post(url, headers=self._headers, json=data)
+        response.raise_for_status()
+        suggestions = response.json()
+        suggestions = [EditSuggestion(**suggestion) for suggestion in suggestions]
+
+        # Add assistant response to history
+        if suggestions:
+            for suggestion in suggestions:
+                self.chat_history.append(
+                    Message(
+                        role=MessageRole.ASSISTANT,
+                        content=suggestion.justification,
+                        timestamp=datetime.now(),
+                    )
+                )
+
+        return suggestions
+
+    def suggest_edits(
+        self,
+        user_query: str,
+        auto_history: bool = True,
+        chat_history: Optional[List[dict]] = None,
+    ) -> List[EditSuggestion]:
+        """Synchronous wrapper for getting edit suggestions."""
+        return self._run_sync(
+            self.asuggest_edits(user_query, auto_history, chat_history)
+        )
+
+    async def await_completion(
+        self, timeout: int = DEFAULT_TIMEOUT, poll_interval: int = DEFAULT_POLL_INTERVAL
+    ) -> Report:
+        """Wait for this report to complete processing."""
+        start_time = time.time()
+        while True:
+            report_response = await self.aget()
+            status = report_response.status
+
+            if status == "completed":
+                return report_response.report
+            elif status == "error":
+                events = await self.aget_events()
+                raise ValueError(f"Report entered error state: {events[-1].msg}")
+            elif time.time() - start_time > timeout:
+                raise TimeoutError(f"Report did not complete within {timeout} seconds")
+
+            await asyncio.sleep(poll_interval)
+
+    def wait_for_completion(
+        self, timeout: int = DEFAULT_TIMEOUT, poll_interval: int = DEFAULT_POLL_INTERVAL
+    ) -> Report:
+        """Synchronous wrapper for awaiting report completion."""
+        return self._run_sync(self.await_completion(timeout, poll_interval))
+
+    async def await_for_plan(
+        self, timeout: int = DEFAULT_TIMEOUT, poll_interval: int = DEFAULT_POLL_INTERVAL
+    ) -> ReportPlan:
+        """Wait for this report's plan to be ready for review."""
+        start_time = time.time()
+        while True:
+            report_metadata = await self.aget_metadata()
+            state = report_metadata.state
+
+            if state == "waiting_approval":
+                report_response = await self.aget()
+                return report_response.plan
+            elif state == "error":
+                events = await self.aget_events()
+                raise ValueError(f"Report entered error state: {events[-1].msg}")
+            elif time.time() - start_time > timeout:
+                raise TimeoutError(f"Plan was not ready within {timeout} seconds")
+
+            await asyncio.sleep(poll_interval)
+
+    def wait_for_plan(
+        self, timeout: int = DEFAULT_TIMEOUT, poll_interval: int = DEFAULT_POLL_INTERVAL
+    ) -> ReportPlan:
+        """Synchronous wrapper for awaiting plan readiness."""
+        return self._run_sync(self.await_for_plan(timeout, poll_interval))
+
+    async def aget_metadata(self) -> ReportMetadata:
+        """Get this report's metadata asynchronously."""
+        return await self._client.aget_report_metadata(self.report_id)
+
+    def get_metadata(self) -> ReportMetadata:
+        """Synchronous wrapper for getting this report's metadata."""
+        return self._run_sync(self.aget_metadata())
+
+    async def adelete(self) -> None:
+        """Delete this report asynchronously."""
+        return await self._client.adelete_report(self.report_id)
+
+    def delete(self) -> None:
+        """Synchronous wrapper for deleting this report."""
+        return self._run_sync(self.adelete())
+
+    async def aaccept_edit(self, suggestion: EditSuggestion) -> None:
+        """Accept a suggested edit.
+
+        Args:
+            suggestion: The EditSuggestion to accept, typically from suggest_edits()
+        """
+        if len(suggestion.blocks) == 0:
+            return
+
+        # Determine if we're editing a plan or report based on first block type
+        is_plan_edit = isinstance(suggestion.blocks[0], ReportPlanBlock)
+
+        # Get current content
+        report_response = await self.aget()
+        current_blocks = (
+            report_response.plan.blocks
+            if is_plan_edit
+            else report_response.report.blocks
+        )
+
+        # Track the edit
+        new_blocks = []
+        for edit_block in suggestion.blocks:
+            # Find matching block in current content
+            old_block = next(
+                (
+                    b
+                    for b in current_blocks
+                    if self._get_block_idx(b) == self._get_block_idx(edit_block)
+                ),
+                None,
+            )
+
+            old_content = (
+                self._get_block_content(old_block) if old_block else "[No old content]"
+            )
+            new_content = self._get_block_content(edit_block)
+
+            if is_plan_edit:
+                new_queries_str = "\n".join(
+                    [
+                        f"Field: {q.field}, Prompt: {q.prompt}, Context: {q.context}"
+                        for q in edit_block.queries
+                    ]
+                )
+                new_dependency_str = (
+                    f"Depends on: {edit_block.dependency}"
+                    if edit_block.dependency
+                    else ""
+                )
+                new_content += f"\n\n{new_queries_str}\n{new_dependency_str}"
+
+                if old_block:
+                    old_queries_str = "\n".join(
+                        [
+                            f"Field: {q.field}, Prompt: {q.prompt}, Context: {q.context}"
+                            for q in old_block.queries
+                        ]
+                    )
+                    old_dependency_str = (
+                        f"Depends on: {old_block.dependency}"
+                        if old_block.dependency
+                        else ""
+                    )
+                    old_content += f"\n\n{old_queries_str}\n{old_dependency_str}"
+
+            self.edit_history.append(
+                EditAction(
+                    block_idx=self._get_block_idx(edit_block),
+                    old_content=old_content,
+                    new_content=new_content,
+                    action="approved",
+                    timestamp=datetime.now(),
+                )
+            )
+
+            # Create updated block
+            if is_plan_edit:
+                new_blocks.append(
+                    ReportPlanBlock(
+                        block=ReportBlock(
+                            idx=edit_block.block.idx,
+                            template=self._get_block_content(edit_block),
+                            sources=edit_block.block.sources,
+                        ),
+                        queries=edit_block.queries,
+                        dependency=edit_block.dependency,
+                    )
+                )
+            else:
+                new_blocks.append(
+                    ReportBlock(
+                        idx=edit_block.idx,
+                        template=self._get_block_content(edit_block),
+                        sources=edit_block.sources,
+                    )
+                )
+
+        if new_blocks:
+            if is_plan_edit:
+                # Update plan in place
+                plan = report_response.plan
+
+                # Replace edited blocks and add new ones
+                for new_block in new_blocks:
+                    block_idx = self._get_block_idx(new_block)
+                    existing_block_idx = next(
+                        (
+                            i
+                            for i, b in enumerate(plan.blocks)
+                            if b.block.idx == block_idx
+                        ),
+                        None,
+                    )
+
+                    if existing_block_idx is not None:
+                        # Replace existing block
+                        plan.blocks[existing_block_idx] = new_block
+                    else:
+                        # Add new block to end
+                        plan.blocks.append(new_block)
+
+                await self.aupdate_plan("edit", plan)
+            else:
+                # Update report in place
+                report = report_response.report
+
+                # Replace edited blocks and add new ones
+                for new_block in new_blocks:
+                    block_idx = self._get_block_idx(new_block)
+                    existing_block_idx = next(
+                        (i for i, b in enumerate(report.blocks) if b.idx == block_idx),
+                        None,
+                    )
+
+                    if existing_block_idx is not None:
+                        # Replace existing block
+                        report.blocks[existing_block_idx] = new_block
+                    else:
+                        # Add new block to end
+                        report.blocks.append(new_block)
+
+                await self.aupdate_report(report)
+
+    def accept_edit(self, suggestion: EditSuggestion) -> None:
+        """Synchronous wrapper for accepting an edit."""
+        return self._run_sync(self.aaccept_edit(suggestion))
+
+    async def areject_edit(self, suggestion: EditSuggestion) -> None:
+        """Reject a suggested edit.
+
+        Args:
+            suggestion: The EditSuggestion to reject, typically from suggest_edits()
+        """
+        # Track the rejections
+        for edit_block in suggestion.blocks:
+            self.edit_history.append(
+                EditAction(
+                    block_idx=self._get_block_idx(edit_block),
+                    old_content=self._get_block_content(edit_block),
+                    new_content=None,
+                    action="rejected",
+                    timestamp=datetime.now(),
+                )
+            )
+
+    def reject_edit(self, suggestion: EditSuggestion) -> None:
+        """Synchronous wrapper for rejecting an edit."""
+        return self._run_sync(self.areject_edit(suggestion))
+
+    def _get_edit_summary_after_message(
+        self, message_timestamp: datetime
+    ) -> Optional[str]:
+        """Get a summary of edits that occurred after a specific message."""
+        relevant_edits = [
+            edit for edit in self.edit_history if edit.timestamp > message_timestamp
+        ]
+
+        if not relevant_edits:
+            return None
+
+        approved = [edit for edit in relevant_edits if edit.action == "approved"]
+        rejected = [edit for edit in relevant_edits if edit.action == "rejected"]
+
+        summary = []
+
+        if approved:
+            summary.append("Approved edits:")
+            for edit in approved:
+                summary.append(
+                    f'Block {edit.block_idx}: "{edit.old_content}" -> "{edit.new_content}"'
+                )
+
+        if rejected:
+            if approved:  # Add spacing if we had approved edits
+                summary.append("")
+            summary.append("Rejected edits:")
+            for edit in rejected:
+                summary.append(f'Block {edit.block_idx}: "{edit.old_content}"')
+
+        return "\n".join(summary)
+
+    async def aget_events(
+        self, last_sequence: Optional[int] = None
+    ) -> List[ReportEventItemEventData_Progress]:
+        """Get all events for this report asynchronously.
+
+        Args:
+            last_sequence: If provided, only get events after this sequence number
+
+        Returns:
+            List of ReportEvent objects
+        """
+        extra_params = []
+        if last_sequence is not None:
+            extra_params.append(f"last_sequence={last_sequence}")
+
+        url = await self._build_url(
+            f"/api/v1/reports/{self.report_id}/events", extra_params
+        )
+
+        response = await self.aclient.get(url, headers=self._headers)
+        response.raise_for_status()
+        progress_events = []
+        for event in response.json():
+            if event["event_type"] == "progress":
+                progress_events.append(
+                    ReportEventItemEventData_Progress(**event["event_data"])
+                )
+
+        return progress_events
+
+    def get_events(
+        self, last_sequence: Optional[int] = None
+    ) -> List[ReportEventItemEventData_Progress]:
+        """Synchronous wrapper for getting report events."""
+        return self._run_sync(self.aget_events(last_sequence))

py/llama_parse/pyproject.toml

@@ -11,13 +11,13 @@ dev = [
 
 [project]
 name = "llama-parse"
-version = "0.6.63"
+version = "0.6.62"
 description = "Parse files into RAG-Optimized formats."
 authors = [{name = "Logan Markewich", email = "logan@llamaindex.ai"}]
 requires-python = ">=3.9,<4.0"
 readme = "README.md"
 license = "MIT"
-dependencies = ["llama-cloud-services>=0.6.63"]
+dependencies = ["llama-cloud-services>=0.6.62"]
 
 [project.scripts]
 llama-parse = "llama_parse.cli.main:parse"

py/pyproject.toml

@@ -19,7 +19,7 @@ dev = [
 
 [project]
 name = "llama-cloud-services"
-version = "0.6.63"
+version = "0.6.62"
 description = "Tailored SDK clients for LlamaCloud services."
 authors = [{name = "Logan Markewich", email = "logan@runllama.ai"}]
 requires-python = ">=3.9,<4.0"

py/tests/classifier/test_client.py

@@ -2,7 +2,6 @@ import os
 import pytest
 from llama_cloud.client import AsyncLlamaCloud
 from llama_cloud.types import Project, ClassifierRule, ClassifyJobResults
-from llama_cloud_services.beta.classifier.types import ClassifyJobResultsWithFiles
 from llama_cloud_services.beta.classifier.client import ClassifyClient
 from llama_cloud_services.files.client import FileClient
 from llama_cloud.errors.unprocessable_entity_error import UnprocessableEntityError
@@ -131,44 +130,6 @@ async def test_classify_file_ids(
         assert item.result.type == expected_type
 
 
-@pytest.mark.asyncio
-async def test_classify_file_ids_from_api_key(
-    e2e_test_settings: EndToEndTestSettings,
-    file_client: FileClient,
-    simple_pdf_file_path: str,
-    research_paper_path: str,
-    classification_rules: list[ClassifierRule],
-):
-    """Test classifying files by their IDs"""
-    # Upload test files first to get their IDs
-    pdf_file = await file_client.upload_file(simple_pdf_file_path)
-    research_paper_file = await file_client.upload_file(research_paper_path)
-
-    classify_client = ClassifyClient.from_api_key(
-        api_key=e2e_test_settings.LLAMA_CLOUD_API_KEY.get_secret_value(),
-        base_url=e2e_test_settings.LLAMA_CLOUD_BASE_URL,
-        project_id=pdf_file.project_id,
-        organization_id=e2e_test_settings.LLAMA_CLOUD_ORGANIZATION_ID,
-    )
-
-    # Classify the uploaded files
-    results = await classify_client.aclassify_file_ids(
-        rules=classification_rules, file_ids=[pdf_file.id, research_paper_file.id]
-    )
-
-    assert isinstance(results, ClassifyJobResults)
-    assert len(results.items) == 2
-
-    file_id_to_expected_type = {
-        pdf_file.id: "number",
-        research_paper_file.id: "research_paper",
-    }
-    # Verify each file got classified
-    for item in results.items:
-        expected_type = file_id_to_expected_type[item.file_id]
-        assert item.result.type == expected_type
-
-
 @parameterize_sync_and_async
 @pytest.mark.asyncio
 async def test_classify_file_path(
@@ -188,7 +149,7 @@ async def test_classify_file_path(
             rules=classification_rules, file_input_path=simple_pdf_file_path
         )
 
-    assert isinstance(results, ClassifyJobResultsWithFiles)
+    assert isinstance(results, ClassifyJobResults)
     assert len(results.items) == 1
 
     # Verify the file got classified
@@ -219,7 +180,7 @@ async def test_classify_file_paths(
             file_input_paths=[simple_pdf_file_path, research_paper_path],
         )
 
-    assert isinstance(results, ClassifyJobResultsWithFiles)
+    assert isinstance(results, ClassifyJobResults)
     assert len(results.items) == 2
 
     file_name_to_expected_type = {
@@ -228,7 +189,8 @@ async def test_classify_file_paths(
     }
     # Verify each file got classified
     for item in results.items:
-        expected_type = file_name_to_expected_type[item.file.name]
+        file = await file_client.get_file(item.file_id)
+        expected_type = file_name_to_expected_type[file.name]
         assert item.result.type == expected_type
 
 

py/tests/report/test_llama_report.py

@@ -0,0 +1,129 @@
+import os
+import pytest
+import uuid
+from typing import AsyncGenerator
+from pytest_asyncio import fixture as async_fixture
+from llama_cloud_services.report import LlamaReport, ReportClient
+
+# Skip tests if no API key is set
+pytestmark = pytest.mark.skipif(
+    not os.getenv("LLAMA_CLOUD_API_KEY") or os.getenv("CI") == "true",
+    reason="No API key provided",
+)
+
+
+@async_fixture(scope="function")
+async def client() -> AsyncGenerator[LlamaReport, None]:
+    """Create a LlamaReport client."""
+    client = LlamaReport()
+    reports_before = await client.alist_reports()
+    reports_before_ids = [r.report_id for r in reports_before]
+    try:
+        yield client
+    finally:
+        # clean up reports
+        try:
+            reports_after = await client.alist_reports()
+            reports_after_ids = [r.report_id for r in reports_after]
+            for report_id in reports_before_ids:
+                if report_id not in reports_after_ids:
+                    await client.adelete_report(report_id)
+        except Exception:
+            pass
+        finally:
+            await client.aclient.aclose()
+
+
+@pytest.fixture(scope="function")
+def unique_name() -> str:
+    """Generate a unique report name."""
+    return f"test-report-{uuid.uuid4()}"
+
+
+@async_fixture(scope="function")
+async def report(
+    client: LlamaReport, unique_name: str
+) -> AsyncGenerator[ReportClient, None]:
+    """Create a report."""
+    report = await client.acreate_report(
+        name=unique_name,
+        template_text=(
+            "# [Some title]\n\n"
+            " ## TLDR\n"
+            "A quick summary of the paper.\n\n"
+            "## Details\n"
+            "More details about the paper, possible more than one section here.\n"
+        ),
+        input_files=["tests/test_files/paper.md"],
+    )
+    try:
+        yield report
+    finally:
+        await report.adelete()
+
+
+@pytest.mark.asyncio
+@pytest.mark.xfail(
+    condition=lambda: os.getenv("CI"),
+    reason="Backend db issues; needs to be fixed.",
+)
+async def test_create_and_delete_report(
+    client: LlamaReport, report: ReportClient
+) -> None:
+    """Test basic report creation and deletion."""
+    # Verify the report exists
+    metadata = await report.aget_metadata()
+    assert metadata.name == report.name
+
+    # Test listing reports
+    reports = await client.alist_reports()
+    assert any(r.report_id == report.report_id for r in reports)
+
+    # Test getting report by ID
+    fetched_report = await client.aget_report(report.report_id)
+    assert fetched_report.report_id == report.report_id
+    assert fetched_report.name == report.name
+
+
+@pytest.mark.asyncio
+@pytest.mark.xfail(
+    condition=lambda: os.getenv("CI"),
+    reason="Report plan sometimes times out",
+    raises=TimeoutError,
+)
+async def test_report_plan_workflow(report: ReportClient) -> None:
+    """Test the report planning workflow."""
+    # Wait for the plan
+    plan = await report.await_for_plan()
+    assert plan is not None
+
+    # Approve the plan
+    response = await report.aupdate_plan(action="approve")
+    assert response is not None
+
+    # Wait for completion
+    completed_report = await report.await_completion()
+    assert len(completed_report.blocks) > 0
+
+    # Get edit suggestions
+    suggestions = await report.asuggest_edits(
+        "TLDR section header more formal.", auto_history=True
+    )
+    assert len(suggestions) > 0
+
+    # Test accepting an edit
+    await report.aaccept_edit(suggestions[0])
+
+    # Get more suggestions and test rejecting
+    more_suggestions = await report.asuggest_edits(
+        "Add a section about machine learning.", auto_history=True
+    )
+    assert len(more_suggestions) > 0
+    await report.areject_edit(more_suggestions[0])
+
+    # Verify chat history is maintained
+    assert len(report.chat_history) >= 4  # 2 user messages + 2 assistant responses
+
+    # get events
+    events = await report.aget_events()
+    assert len(events) > 0

py/uv.lock

@@ -1596,7 +1596,7 @@ wheels = [
 
 [[package]]
 name = "llama-cloud-services"
-version = "0.6.63"
+version = "0.6.61"
 source = { editable = "." }
 dependencies = [
     { name = "click", version = "8.1.8", source = { registry = "https://pypi.org/simple" }, marker = "python_full_version < '3.10'" },

report.md

@@ -0,0 +1,101 @@
+# LlamaReport (beta/invite-only)
+
+LlamaReport is a prebuilt agentic report builder that can be used to build reports from a variety of data sources.
+
+The python SDK for interacting with the LlamaReport API. The SDK provides two main classes:
+
+- `LlamaReport`: For managing reports (create, list, delete)
+- `ReportClient`: For working with a specific report (editing, approving, etc.)
+
+## Quickstart
+
+```bash
+pip install llama-cloud-services
+```
+
+```python
+from llama_cloud_services import LlamaReport
+
+# Initialize the client
+client = LlamaReport(
+    api_key="your-api-key",
+    # Optional: Specify project_id, organization_id, async_httpx_client
+)
+
+# Create a new report
+report = client.create_report(
+    "My Report",
+    # must have one of template_text or template_instructions
+    template_text="Your template text",
+    template_instructions="Instructions for the template",
+    # must have one of input_files or retriever_id
+    input_files=["data1.pdf", "data2.pdf"],
+    retriever_id="retriever-id",
+)
+```
+
+## Working with Reports
+
+The typical workflow for a report involves:
+
+1. Creating the report
+2. Waiting for and approving the plan
+3. Waiting for report generation
+4. Making edits to the report
+
+Here's a complete example:
+
+```python
+# Create a report
+report = client.create_report(
+    "Quarterly Analysis", input_files=["q1_data.pdf", "q2_data.pdf"]
+)
+
+# Wait for the plan to be ready
+plan = report.wait_for_plan()
+
+# Option 1: Directly approve the plan
+report.update_plan(action="approve")
+
+# Option 2: Suggest and review edits to the plan
+suggestions = report.suggest_edits(
+    "Can you add a section about market trends?"
+)
+for suggestion in suggestions:
+    print(suggestion)
+
+    # Accept or reject the suggestion
+    if input("Accept? (y/n): ").lower() == "y":
+        report.accept_edit(suggestion)
+    else:
+        report.reject_edit(suggestion)
+
+# Wait for the report to complete
+report = report.wait_for_completion()
+
+# Make edits to the final report
+suggestions = report.suggest_edits("Make the executive summary more concise")
+
+# Review and accept/reject suggestions as above
+...
+```
+
+### Getting the Final Report
+
+Once you are satisfied with the report, you can get the final report object and use the content as you see fit.
+
+Here's an example of printing out the final report:
+
+```python
+report = report.get()
+report_text = "\n\n".join([block.template for block in report.blocks])
+
+print(report_text)
+```
+
+## Additional Features
+
+- **Async Support**: All methods have async counterparts: `create_report` -> `acreate_report`, `wait_for_plan` -> `await_for_plan`, etc.
+- **Automatic Chat History**: The SDK automatically keeps track of chat history for each suggestion, unless you specify `auto_history=False` in `suggest_edits`.
+- **Custom HTTP Client**: You can provide your own `httpx.AsyncClient` to the `LlamaReport` class.
+- **Project and Organization IDs**: You can specify `project_id` and `organization_id` to use a specific project or organization.

scripts/version-bump.py

@@ -8,11 +8,10 @@ import subprocess
 import sys
 import tomlkit
 from pathlib import Path
-import json
 
 
-def get_current_versions() -> tuple[str, str, str, str | None]:
-    """Get current versions from both pyproject.toml files and TS package.json."""
+def get_current_versions() -> tuple[str, str, str]:
+    """Get current versions from both pyproject.toml files."""
     # Read main pyproject.toml
     main_content = Path("py/pyproject.toml").read_text()
     main_doc = tomlkit.parse(main_content)
@@ -35,21 +34,11 @@ def get_current_versions() -> tuple[str, str, str, str | None]:
             )
             break
 
-    # Read TypeScript package.json version via helper
-    ts_version: str = get_ts_version()
-
-    return (
-        str(main_version),
-        str(llama_parse_version),
-        str(dependency_version),
-        str(ts_version) if ts_version is not None else None,
-    )
+    return str(main_version), str(llama_parse_version), str(dependency_version)
 
 
 def validate_versions(
-    main_version: str,
-    llama_parse_version: str,
-    dependency_version: str,
+    main_version: str, llama_parse_version: str, dependency_version: str
 ) -> list[str]:
     """Validate that versions are consistent and return warnings."""
     warnings = []
@@ -71,7 +60,7 @@ def validate_versions(
 
 
 def set_version(version: str) -> None:
-    """Set version across Python projects (no TS change)."""
+    """Set version across all pyproject.toml files using tomlkit to preserve formatting."""
     # Update main pyproject.toml
     main_content = Path("py/pyproject.toml").read_text()
     main_doc = tomlkit.parse(main_content)
@@ -90,26 +79,7 @@ def set_version(version: str) -> None:
             break
     Path("py/llama_parse/pyproject.toml").write_text(tomlkit.dumps(llama_parse_doc))
 
-    click.echo(f"Updated Python versions to {version}")
-
-
-def get_ts_version() -> str:
-    """Read TypeScript package.json version (if present)."""
-    ts_package_path = Path("ts/llama_cloud_services/package.json")
-    package_data = json.loads(ts_package_path.read_text())
-    data = package_data.get("version")
-    if data is None:
-        raise RuntimeError("TypeScript package.json version not found")
-    return data
-
-
-def set_ts_version(version: str) -> None:
-    """Set TypeScript package.json version only."""
-    ts_package_path = Path("ts/llama_cloud_services/package.json")
-    package_data = json.loads(ts_package_path.read_text())
-    package_data["version"] = version
-    ts_package_path.write_text(json.dumps(package_data, indent=2) + "\n")
-    click.echo(f"Updated TypeScript package.json version to {version}")
+    click.echo(f"Updated all versions to {version}")
 
 
 def get_current_branch() -> str:
@@ -129,7 +99,7 @@ def create_if_not_exists(version: str) -> None:
         )
         sys.exit(1)
 
-    tag_name = f"v{version}" if version[0].isdigit() else version
+    tag_name = f"v{version}"
     if not tag_exists(version):
         # Create tag
         subprocess.run(["git", "tag", tag_name], check=True)
@@ -163,18 +133,12 @@ def cli() -> None:
 @cli.command()
 def get() -> None:
     """Get current versions and show validation warnings."""
-    (
-        main_version,
-        llama_parse_version,
-        dependency_version,
-        ts_version,
-    ) = get_current_versions()
+    main_version, llama_parse_version, dependency_version = get_current_versions()
 
     click.echo("Current versions:")
     click.echo(f"  llama-cloud-services: {main_version}")
     click.echo(f"  llama-parse: {llama_parse_version}")
     click.echo(f"  dependency reference: {dependency_version}")
-    click.echo(f"  typescript package: {ts_version}")
 
     warnings = validate_versions(main_version, llama_parse_version, dependency_version)
     if warnings:
@@ -187,15 +151,9 @@ def get() -> None:
 
 @cli.command()
 @click.argument("version")
-@click.option("--js", is_flag=True, help="Update TypeScript package.json only")
-def set(version: str, js: bool) -> None:
-    """Set version for Python, TypeScript, or both (default: Python only)."""
-
-    if js:
-        set_ts_version(version)
-        return
-    else:
-        set_version(version)
+def set(version: str) -> None:
+    """Set version across all pyproject.toml files."""
+    set_version(version)
 
 
 @cli.command()
@@ -207,16 +165,11 @@ def set(version: str, js: bool) -> None:
     is_flag=True,
     help="Push the tag to the remote repository",
 )
-@click.option(
-    "--js",
-    is_flag=True,
-    help="tag TypeScript package.json only",
-)
-def tag(version: str | None = None, push: bool = False, js: bool = False) -> None:
+def tag(version: str | None = None, push: bool = False) -> None:
     """Create and push a git tag for the current version."""
     if not version:
-        main_version, _, _, js_version = get_current_versions()
-        version = f"llama-cloud-services@{js_version}" if js else main_version
+        main_version, _, _ = get_current_versions()
+        version = main_version
 
     create_if_not_exists(version)
     if push:

ts/llama_cloud_services/package.json

@@ -1,6 +1,6 @@
 {
   "name": "llama-cloud-services",
-  "version": "0.3.4",
+  "version": "0.3.3",
   "type": "module",
   "license": "MIT",
   "scripts": {
@@ -9,7 +9,6 @@
     "dev": "bunchee --watch",
     "lint": "eslint src/ --ignore-pattern client/*.ts --no-warn-ignored",
     "format": "prettier --write ./src/",
-    "format:check": "prettier --check ./src/",
     "test": "vitest run --testTimeout=60000",
     "test:watch": "vitest --watch",
     "test:ui": "vitest --ui",

ts/llama_cloud_services/src/beta/agent/client.ts

@@ -1,5 +1,5 @@
-import { createClient } from "@hey-api/client-fetch";
-import { client as defaultClient } from "../../api";
+import { createClient, createConfig } from "@hey-api/client-fetch";
+import { getEnv } from "@llamaindex/env";
 import {
   aggregateAgentDataApiV1BetaAgentDataAggregatePost,
   createAgentDataApiV1BetaAgentDataPost,
@@ -24,19 +24,36 @@ import type {
  */
 export class AgentClient<T = unknown> {
   private client: ReturnType<typeof createClient>;
+  private baseUrl: string;
+  private headers: Record<string, string>;
   private collection: string;
   private agentUrlId: string;
 
   constructor({
-    client = defaultClient,
+    apiKey = getEnv("LLAMA_CLOUD_API_KEY"),
+    baseUrl = "https://api.cloud.llamaindex.ai/",
     collection = "default",
     agentUrlId = "_public",
   }: {
-    client?: ReturnType<typeof createClient>;
+    apiKey?: string;
+    baseUrl?: string;
     collection?: string;
     agentUrlId?: string;
   }) {
-    this.client = client;
+    this.baseUrl = baseUrl;
+
+    this.headers = {
+      "X-SDK-Name": "llamaindex-ts",
+      ...(apiKey && { Authorization: `Bearer ${apiKey}` }),
+    };
+
+    this.client = createClient(
+      createConfig({
+        baseUrl: this.baseUrl,
+        headers: this.headers,
+      }),
+    );
+
     this.collection = collection;
     this.agentUrlId = agentUrlId;
   }
@@ -264,13 +281,15 @@ export interface AgentDataClientOptions {
  * @returns A new AgentClient instance
  */
 export function createAgentDataClient<T = unknown>({
-  client = defaultClient,
+  apiKey,
+  baseUrl,
   windowUrl,
   env,
   agentUrlId,
   collection = "default",
 }: {
-  client?: ReturnType<typeof createClient>;
+  apiKey?: string;
+  baseUrl?: string;
   windowUrl?: string;
   env?: Record<string, string>;
   agentUrlId?: string;
@@ -302,8 +321,9 @@ export function createAgentDataClient<T = unknown>({
   }
 
   return new AgentClient({
+    ...(apiKey && { apiKey }),
+    ...(baseUrl && { baseUrl }),
     ...(agentUrlId && { agentUrlId }),
     collection,
-    client,
   });
 }

ts/llama_cloud_services/src/reader.ts

@@ -194,7 +194,7 @@ export class LlamaParseReader extends FileReader {
       ? this.language
       : [this.language];
     this.stdout =
-      (params.stdout ?? typeof process !== "undefined")
+      params.stdout ?? typeof process !== "undefined"
         ? process!.stdout
         : undefined;
     const apiKey = params.apiKey ?? getEnv("LLAMA_CLOUD_API_KEY");

ts/llama_cloud_services/tests/client.test.ts

@@ -1,80 +0,0 @@
-import { describe, it, expect, beforeEach, afterEach, vi } from "vitest";
-import { listProjectsApiV1ProjectsGet, client } from "../src/api.js";
-
-describe("Global client configuration", () => {
-  const originalFetch = globalThis.fetch;
-
-  beforeEach(() => {
-    vi.restoreAllMocks();
-  });
-
-  afterEach(() => {
-    globalThis.fetch = originalFetch;
-  });
-
-  it("adds X-SDK-Name header from global client config", async () => {
-    const fetchSpy = vi
-      .spyOn(globalThis, "fetch")
-      .mockImplementation(async (input, init) => {
-        // Validate the header is present on the outgoing request
-        let headers: Headers;
-        if (input && typeof input === "object" && "headers" in (input as any)) {
-          headers = (input as Request).headers;
-        } else {
-          headers = new Headers((init && init.headers) || {});
-        }
-        expect(headers.get("X-SDK-Name")).toBe("llamaindex-ts");
-
-        return new Response(JSON.stringify([]), {
-          status: 200,
-          headers: { "Content-Type": "application/json" },
-        });
-      });
-
-    // Trigger any request via the generated SDK (imported through src/api.ts)
-    await listProjectsApiV1ProjectsGet({ throwOnError: false });
-
-    expect(fetchSpy).toHaveBeenCalledOnce();
-  });
-
-  it("respects additional custom headers set via setConfig", async () => {
-    const prevConfig = client.getConfig();
-    try {
-      client.setConfig({
-        ...prevConfig,
-        headers: {
-          ...(prevConfig.headers || {}),
-          "X-Custom-Header": "custom-value",
-        },
-      });
-
-      const fetchSpy = vi
-        .spyOn(globalThis, "fetch")
-        .mockImplementation(async (input, init) => {
-          let headers: Headers;
-          if (
-            input &&
-            typeof input === "object" &&
-            "headers" in (input as any)
-          ) {
-            headers = (input as Request).headers;
-          } else {
-            headers = new Headers((init && init.headers) || {});
-          }
-          expect(headers.get("X-SDK-Name")).toBe("llamaindex-ts");
-          expect(headers.get("X-Custom-Header")).toBe("custom-value");
-
-          return new Response(JSON.stringify([]), {
-            status: 200,
-            headers: { "Content-Type": "application/json" },
-          });
-        });
-
-      await listProjectsApiV1ProjectsGet({ throwOnError: false });
-      expect(fetchSpy).toHaveBeenCalledOnce();
-    } finally {
-      // Restore original configuration to avoid test cross-talk
-      client.setConfig(prevConfig);
-    }
-  });
-});