fix(memex): add Feishu OAuth support, fix async deadlock, and adapt for local dev by A11en0 · Pull Request #237 · volcengine/OpenViking

A11en0 · 2026-02-20T17:56:08Z

Description

Enhance the Memex Feishu integration with full OAuth support, MCP protocol compliance, and fix several issues for local development with OpenAI provider. Also fix a re-entrant deadlock in run_async() that blocked session-aware search.

Type of Change

Bug fix (non-breaking change that fixes an issue)
New feature (non-breaking change that adds functionality)
Documentation update

Changes Made

Feishu Integration Enhancements (`feishu.py`)

Add MCP protocol handshake (initialize + notifications/initialized) on server start
Fix tool names to use underscores (docx_v1_document_rawContent) and nested path/data/params arguments
Add auto-incrementing request IDs and skip notification messages in response parsing
Parse MCP content[0].text JSON wrapper in call_tool() responses
Add full OAuth user login flow (/feishu-login) with local callback server and token refresh
Add list_user_files() for browsing user's drive / My Space
Add search_wiki() for wiki node search
Rewrite search_documents() to use REST API directly (works with both user and tenant tokens)
Add new CLI commands: /feishu-login, /feishu-ls, /feishu-list, /feishu-tools

Async Deadlock Fix (`openviking_cli/utils/async_utils.py`)

Make run_async() re-entrant safe: detect running event loop via asyncio.get_running_loop() and spawn a temporary thread+loop for nested coroutines
Root cause: Session.get_context_for_search() calls run_async() from within code already running on the shared background event loop, causing the loop thread to block waiting on itself

Local Dev Fixes

Fix VLM backend config: fallback from backend → provider field → "openai" default (config.py)
Restore client.search() with session support for multi-type retrieval (memory + resource + skill) (rag/recipe.py)
Add [tool.uv.sources] for local editable openviking install (pyproject.toml)
Change requires-python from >=3.13 to >=3.10 for broader compatibility
Switch build system entry point to memex.cli:main with setuptools
Replace deleted openviking.utils.logger monkey-patch with standard logging module (cli.py)
Convert all imports to relative imports for package-style execution

Documentation (`README.md`)

Add missing commands: /stat, /glob, /session, /commit, /memories, /feishu-search, /feishu-tools
Add OpenAI provider configuration example alongside volcengine
Add Feishu OAuth setup instructions (permissions, redirect URL)
Fix step numbering, project structure, and test command path

Testing

I have tested this on the following platforms:
- Linux

Checklist

My code follows the project's coding style
I have performed a self-review of my code
I have commented my code, particularly in hard-to-understand areas
I have made corresponding changes to the documentation
My changes generate no new warnings

- CLI-based personal knowledge assistant powered by OpenViking - Knowledge management: add files, directories, URLs - Intelligent Q&A with RAG and multi-turn conversation - Knowledge browsing with L0/L1/L2 context layers - Semantic search with intent analysis - Feishu/Lark MCP integration for document import - Rich CLI interface with prompt_toolkit

- Fix /tree command to handle dict list return type - Fix RAG recipe to load LLM config from ov.conf instead of removed attributes - Fix find/search methods to use 'limit' parameter instead of 'top_k'

- Strip trailing slashes from URIs to prevent path errors - Improve tree display to handle dict list return type

- Add recursive directory import with file type filtering - Improve error handling for add_resource results - Add proper cleanup on exit to kill AGFS processes - Fix import_dir to use new directory handling logic

…ll content - Lower default score_threshold from 0.3 to 0.1 (actual scores are 0.17-0.25) - Fix build_context to use client.read() for full content instead of just abstract - This fixes the issue where queries returned no results despite data being indexed

…, smart context loading

…ction, hybrid search, rerank - P0: Integrate OpenViking Session for context-aware search with IntentAnalyzer - P1: Add session.commit() for long-term memory extraction (6 categories) - P2: Implement hybrid search strategy (find for simple, search for complex) - P3: Add LLM-based rerank for improved result relevance New CLI commands: /session, /commit, /memories

- abstract() only works for directories, not files - Changed to try read() first, then abstract() for directories - This fixes the issue where only 1 result was returned instead of 5

- Remove high_confidence_threshold and low_confidence_threshold - Remove rerank (use OpenViking score directly) - Remove confidence hints from system prompt - Unified content loading: all results get full content (2000 chars) - Simpler build_context without confidence logic

…xpansion - Remove _should_use_deep_search and find() auto-selection - Always use search() with session for IntentAnalyzer - Add _expand_query() to expand Chinese queries with English keywords - Example: 'OpenViking 怎么安装' → 'OpenViking 怎么安装 installation setup install pip' - Installation doc now ranks #1 (was not in top 5 before)

… query expansion

- Add wait_processed() after session commit for memory embedding - Add OpenAI api_base default (https://api.openai.com/v1) - Remove [:2000] content truncation that caused unexpected results

- Add -v/--verbose flag to control logging (default: off) - Fix Quick Start: add cd examples/memex step - Update CLI Options to match actual parameters

…x' works pyproject.toml was missing [build-system], causing uv to skip entry point installation. Also fixed entry point from 'cli:main' to 'memex.cli:main' and converted all bare imports to relative imports. Updated README to match actual project structure.

openviking creates per-module loggers with their own handlers and propagate=False, making root logger level ineffective. Patch get_logger to enforce WARNING level for all current and future openviking loggers.

- Implement OAuth 2.0 authorization code flow for user access token - Add support for manual code entry (for headless/SSH environments) - Fix tool names (underscore format) and parameter structure (nested) - Replace MCP file listing with REST API search (bypassing MCP limitations) - Add /feishu-login and improve /feishu-list commands

- Implement /feishu-ls using drive.v1.file.list API - Allows browsing user's 'My Space' (root) and subfolders - Requires user access token via /feishu-login - Updated docs with redirect URI configuration step

# Conflicts: # examples/memex/README.md # examples/memex/cli.py # examples/memex/client.py # examples/memex/commands/stats.py # examples/memex/pyproject.toml

- Fix VLM backend config fallback (provider field support) - Use find() instead of search() to avoid IntentAnalyzer deadlock - Add uv source override for local openviking editable install - Replace deleted logger monkey-patch with standard logging - Update README: add missing commands, OpenAI config example, fix structure

Session methods call run_async() from within code already running on the shared background event loop, causing a deadlock when search() is used with session context. Fix by detecting a running event loop via asyncio.get_running_loop() and spawning a temporary thread+loop for the nested coroutine.

CLAassistant · 2026-02-20T17:56:15Z

Thank you for your submission! We really appreciate it. Like many open source projects, we ask that you sign our Contributor License Agreement before we can accept your contribution.
_{You have signed the CLA already but the status is still pending? Let us recheck it.}

A11en0 added 25 commits February 5, 2026 04:05

fix: fix Memex bugs in tree display and RAG config loading

f42d486

- Fix /tree command to handle dict list return type - Fix RAG recipe to load LLM config from ov.conf instead of removed attributes - Fix find/search methods to use 'limit' parameter instead of 'top_k'

fix: fix URI path handling in abstract/overview commands

6eb44ee

- Strip trailing slashes from URIs to prevent path errors - Improve tree display to handle dict list return type

feat: improve Memex directory support and cleanup

c23af6c

- Add recursive directory import with file type filtering - Improve error handling for add_resource results - Add proper cleanup on exit to kill AGFS processes - Fix import_dir to use new directory handling logic

feat: add Tier 1 RAG improvements - query rewriting, confidence hints…

e2b07bf

…, smart context loading

fix: fix content retrieval in _process_search_results

3089593

- abstract() only works for directories, not files - Changed to try read() first, then abstract() for directories - This fixes the issue where only 1 result was returned instead of 5

Merge branch 'volcengine:main' into feature/new_demo

57717e9

docs: update memex README with architecture details, remove redundant…

20a1cb8

… query expansion

fix: address PR review feedback

bde395f

- Add wait_processed() after session commit for memory embedding - Add OpenAI api_base default (https://api.openai.com/v1) - Remove [:2000] content truncation that caused unexpected results

Merge branch 'volcengine:main' into feature/new_demo

0d153c1

feat: add --verbose flag, fix README quick start

fa375ee

- Add -v/--verbose flag to control logging (default: off) - Fix Quick Start: add cd examples/memex step - Update CLI Options to match actual parameters

fix(memex): suppress openviking INFO logs in non-verbose mode

63090ce

openviking creates per-module loggers with their own handlers and propagate=False, making root logger level ineffective. Patch get_logger to enforce WARNING level for all current and future openviking loggers.

feat(memex): add /feishu-ls command for directory browsing

905035d

- Implement /feishu-ls using drive.v1.file.list API - Allows browsing user's 'My Space' (root) and subfolders - Requires user access token via /feishu-login - Updated docs with redirect URI configuration step

fix(memex): auto-resolve root folder token when folder_token is empty

d42eb09

Merge remote-tracking branch 'origin/main' into feature/new_demo

bd1e41c

# Conflicts: # examples/memex/README.md # examples/memex/cli.py # examples/memex/client.py # examples/memex/commands/stats.py # examples/memex/pyproject.toml

refactor(memex): switch back to client.search() for multi-type retrieval

753b99c

Merge remote-tracking branch 'upstream/main' into feature/new_demo

0493995

github-project-automation bot added this to OpenViking project Feb 20, 2026

github-project-automation bot moved this to Backlog in OpenViking project Feb 20, 2026

MaojiaSheng approved these changes Feb 21, 2026

View reviewed changes

MaojiaSheng merged commit a5ead81 into volcengine:main Feb 21, 2026
19 of 21 checks passed

github-project-automation bot moved this from Backlog to Done in OpenViking project Feb 21, 2026

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

fix(memex): add Feishu OAuth support, fix async deadlock, and adapt for local dev#237

fix(memex): add Feishu OAuth support, fix async deadlock, and adapt for local dev#237
MaojiaSheng merged 25 commits intovolcengine:mainfrom
A11en0:feature/new_demo

A11en0 commented Feb 20, 2026

Uh oh!

CLAassistant commented Feb 20, 2026

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants

Conversation

A11en0 commented Feb 20, 2026

Description

Type of Change

Changes Made

Feishu Integration Enhancements (feishu.py)

Async Deadlock Fix (openviking_cli/utils/async_utils.py)

Local Dev Fixes

Documentation (README.md)

Testing

Checklist

Uh oh!

CLAassistant commented Feb 20, 2026

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants

Feishu Integration Enhancements (`feishu.py`)

Async Deadlock Fix (`openviking_cli/utils/async_utils.py`)

Documentation (`README.md`)