mirror of
https://github.com/langgenius/dify-plugin-sdks.git
synced 2026-07-22 10:25:23 -04:00
Maries
c6f83a63e1
* chore: fix ruff issue * feat(oauth): implement OAuth * feat(invoke-message): refactor message handling and introduce InvokeMessage class * feat(plugin-oauth): add credential_id and credential_type to tool parameters * feat(plugin-oauth): add credential_id and credential_type to tool parameters * chore: update dify_plugin version to 0.5.0b4 and clean up github.yaml * chore: update plugin version to 0.1.2 in manifest.yaml * feat(session): session context and tool backwards invocation credential support * feat(oauth): session context and tool backwards invocation credential support * feat: update README and requirements for OAuth support in version 0.4.2 * feat: add .gitignore to exclude IDE files and secret keys * chore: apply ruff * feat: bump version to 0.4.2b1 * feat: update GitHub plugin configuration for OAuth support and improve credential handling * feat: update .gitignore to exclude dify plugin files and public keys * feat: fix credential validation for GitHub API and bump version to 0.2.1 * feat: update GitHub plugin to support multiple access tokens and bump version to 0.2.5 * chore: apply ruff * feat: add ToolProviderOAuthError for improved OAuth error handling in GitHub plugin * chore: apply ruff * chore: bump version to 0.4.2 * chore: update examples sdk version to 0.4.2 * fix: thread deadlock in PluginRunner when running tests without gevent monkey patching * feat: add support for refreshing OAuth credentials in Plugin and GitHub provider * feat: refactor OAuth credential handling to return structured OAuthCredentials object * apply ruff * feat: refactor OAuth credential handling to use ToolOAuthCredentials for improved structure * feat: reorganize imports in __init__.py for improved clarity and structure * feat: add Microsoft To Do plugin for refresh token example * chore: apply ruff * fix: update author in GitHub configuration and clean up Microsoft To Do schema * chore: bump version to 0.4.2b2 in pyproject.toml * feat: update Microsoft To Do plugin to handle OAuth token encoding and version bump * feat:remove inelegant example * chore: update dify_plugin version to 0.4.2 * chore: bump version to 0.4.2 in pyproject.toml --------- Co-authored-by: Yeuoly <admin@srmxy.cn>
Overview
The Agent node in Dify Chatflow/Workflow lets LLMs autonomously use tools. This plugin features two official Dify Agent reasoning strategies, enabling LLMs to dynamically select and run tools during runtime for multi-step problem-solving.
Strategies
1. Function Calling
Function Calling maps user commands to specific functions or tools. The LLM identifies the user's intent, decides which function to call, and extracts the required parameters. It is a straightforward mechanism for invoking external capabilities.
Pros:
- Precise: Directly calls the right tool for defined tasks, avoiding complex reasoning.
- Easy External Integration: Integrates external APIs and tools as callable functions.
- Structured Output: Provides structured function call information for easy processing.
2. ReAct (Reason + Act)
ReAct alternates between the LLM reasoning about the situation and taking actions. The LLM analyzes the current state and goal, selects and uses a tool, and then uses the tool's output for the next thought and action. This cycle repeats until the problem is resolved.
Pros:
- Leverages External Information: Effectively uses external tools to gather information for tasks the model cannot handle alone.
- Explainable Reasoning: Interwoven reasoning and action steps allow some tracking of the Agent's process.
- Wide Applicability: Suitable for tasks requiring external knowledge or specific actions, such as Q&A, information retrieval, and task execution.