openclaw-knowledge-distiller
Provides a Python CLI (`kd`) and an MCP server to convert video URLs (YouTube/Bilibili/Facebook) into structured knowledge outputs by extracting subtitles when available, otherwise performing local ASR (Qwen3-ASR on Apple Silicon), and optionally generating summaries via external LLM providers (Google/OpenAI/Anthropic).
Score Breakdown
⚙ Agent Friendliness
🔒 Security
Uses environment/config for provider API key; exact secret handling and logging behavior is not documented. MCP server runs over stdio (no network endpoint implied), but execution involves fetching media (yt-dlp) and calling external LLM providers when summarization is enabled. Rate limiting and operational security controls are not described.
⚡ Reliability
Best When
You have Apple Silicon and want local transcription from videos, optionally followed by LLM summarization, and you want an MCP-based agent workflow (submit URL → poll status → fetch result).
Avoid When
You need a publicly hosted web API, strict security controls (auditing, guaranteed secret handling), or comprehensive documentation of rate limits/retry behavior for robust automation.
Use Cases
- • Programmatically distill video content into structured notes for research/study
- • Use MCP tooling (Claude Code / Open CLAW) to run video-to-knowledge jobs with polling and result retrieval
- • Offline/zero-cloud transcription on Apple Silicon with optional AI summarization
Not For
- • Server-side multi-tenant use where security/compliance guarantees are required
- • Environments without macOS Apple Silicon (for the advertised local Qwen3-ASR path)
- • High-scale production pipelines needing documented SLAs, stable APIs, and strong error/retry semantics
Interface
Authentication
No explicit OAuth scopes described. For summarization providers, a single API key is configured; local transcription mode (`--no-summary`) avoids external provider auth.
Pricing
README markets local ASR as free/zero key; external summarization providers would be paid by the user.
Agent Metadata
Known Gotchas
- ⚠ Long-running jobs require polling (`get_status`) until completion.
- ⚠ Subtitle availability affects behavior/performance (subtitle extraction vs ASR).
- ⚠ MCP tool outputs depend on chosen `format`/result phase; agents should request `get_result` with correct format.
- ⚠ External summarization requires correct provider/model configuration; local-only mode avoids provider keys.
Alternatives
Full Evaluation Report
Comprehensive deep-dive: security analysis, reliability audit, agent experience review, cost modeling, competitive positioning, and improvement roadmap for openclaw-knowledge-distiller.
AI-powered analysis · PDF + markdown · Delivered within 30 minutes
Package Brief
Quick verdict, integration guide, cost projections, gotchas with workarounds, and alternatives comparison.
Delivered within 10 minutes
Score Monitoring
Get alerted when this package's AF, security, or reliability scores change significantly. Stay ahead of regressions.
Continuous monitoring
Scores are editorial opinions as of 2026-03-30.