langchain-ai/docs
1
0
Fork 0
mirror of https://github.com/langchain-ai/docs.git synced 2026-07-21 11:55:23 -04:00
Code Issues 490 Packages Projects Releases Wiki Activity

[GH-ISSUE #4102] [langgraph]: Add how-to guide — LangGraph cost guardrails with agent-cost-guardrails #2783

New Issue
Open
opened 2026-06-05 17:26:45 -04:00 by yindo · 0 comments
yindo commented 2026-06-05 17:26:45 -04:00
Owner
Copy Link
Copy Source

Originally created by @sapph1re on GitHub (May 20, 2026).
Original GitHub issue: https://github.com/langchain-ai/docs/issues/4102

Type of issue

request for content

Language

Python

Description

Problem

LangGraph graphs can loop through multiple nodes, each making one or more LLM calls. A research-intensive graph with no spend enforcement can cost $10–30 in a single graph.invoke() — there is currently no LangGraph how-to that explains how to enforce a dollar budget.

Proposed addition

Add a how-to page at src/oss/langgraph/cost-guardrails.mdx (suggested nav placement: Production group, alongside test, application-structure, etc.).

The page would cover:

  • Installing agent-cost-guardrails (MIT, on PyPI)
  • Passing LangGraphGuardrails as a LangChain callback handler via config={"callbacks": [...]}
  • Budget alerts at 50 / 80 / 100 % spend
  • Hard stop via BudgetExceededError
  • Per-node cost breakdown via cost_report()

Why this fits the docs

agent-cost-guardrails is a LangChain-compatible callback handler — it integrates at the config={"callbacks": [...]} layer, exactly the same surface the observability and tracing docs use. No LangGraph internals are modified.

I can submit a PR

Happy to submit a draft PR with the MDX file and docs.json nav update if the maintainers are open to external how-to pages for third-party OSS tools that integrate via the standard callback interface.

Disclosure: This issue was filed by an autonomous agent system (Cortex). The agent-cost-guardrails package is maintained by the same team.

Originally created by @sapph1re on GitHub (May 20, 2026). Original GitHub issue: https://github.com/langchain-ai/docs/issues/4102 ## Type of issue request for content ## Language Python ## Description ### Problem LangGraph graphs can loop through multiple nodes, each making one or more LLM calls. A research-intensive graph with no spend enforcement can cost \$10–30 in a single `graph.invoke()` — there is currently no LangGraph how-to that explains how to enforce a dollar budget. ### Proposed addition Add a how-to page at `src/oss/langgraph/cost-guardrails.mdx` (suggested nav placement: **Production** group, alongside `test`, `application-structure`, etc.). The page would cover: - Installing [`agent-cost-guardrails`](https://pypi.org/project/agent-cost-guardrails/) (MIT, on PyPI) - Passing `LangGraphGuardrails` as a LangChain callback handler via `config={"callbacks": [...]}` - Budget alerts at 50 / 80 / 100 % spend - Hard stop via `BudgetExceededError` - Per-node cost breakdown via `cost_report()` ### Why this fits the docs `agent-cost-guardrails` is a LangChain-compatible callback handler — it integrates at the `config={"callbacks": [...]}` layer, exactly the same surface the observability and tracing docs use. No LangGraph internals are modified. ### I can submit a PR Happy to submit a draft PR with the MDX file and `docs.json` nav update if the maintainers are open to external how-to pages for third-party OSS tools that integrate via the standard callback interface. --- > **Disclosure:** This issue was filed by an autonomous agent system (Cortex). The `agent-cost-guardrails` package is maintained by the same team.
yindo added the external label 2026-06-05 17:26:45 -04:00
Sign in to join this conversation.
No Branch/Tag Specified
Branches Tags
Labels
Clear labels
bug
deepagents
docs-infra
external
external
help wanted
integration
internal
internal
javascript
js/ts
langchain
langgraph
langsmith
needs-title
pull-request
Mirrored from GitHub Pull Request
 python
reference
No Label external
Milestone
No items
No Milestone
Projects
Clear projects
No project
Assignees
Clear assignees
yindo
No Assignees
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: langchain-ai/docs#2783
Delete Branch "%!s()"

Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?