技能列表

Extract text from images using Tesseract.js OCR (100% local, no API key required). Supports Chinese (simplified/traditional) and English.

OCR document extraction - extract text from scanned documents, photos, and images using OCR. Use when reading scanned PDFs, photographed pages, handwritten notes, or any document that needs optical character recognition.

|

Build a searchable local knowledge base from an Obsidian vault's "笔记同步助手" inbox, then answer with citations, topic cards, update logs, and daily digests for OpenClaw. Use when users want to structure synced Obsidian notes into a retrieval-ready knowledge base.

Semantic search for Obsidian notes using AI vector embeddings. Find notes by meaning, discover connections, filter by tags/dates/folders, and retrieve full content. Use when the user says 'search my notes', 'find in Obsidian', 'what did I write about', 'related notes', 'Obsidian search', or wants AI-powered knowledge retrieval from their Obsidian vault.

Create and develop Obsidian plugins from scratch. Use when building a new Obsidian plugin, scaffolding from the sample-plugin-plus template, or developing plugin features. Covers project setup, manifest configuration, TypeScript development, settings UI, commands, ribbons, modals, and Obsidian API patterns.

Structured persistent memory system using an Obsidian vault with daily journals, project docs, knowledge base, self-improvement logging, and Discord workspace integration. Use when: (1) Setting up a new OpenClaw agent's memory system, (2) Agent needs persistent memory across sessions, (3) Organizing project documentation, daily journals, or knowledge base, (4) Logging errors, learnings, or feature requests for continuous improvement, (5) Setting up Discord as primary workspace (voice, components

|

Control a remote Obsidian vault through Fast Note Sync. Use when reading, searching, writing, or appending notes in Obsidian from OpenClaw, especially for remote vault workflows that do not have direct filesystem access.

Deep SLO/SLI workflow—user-centric SLIs, SLO targets and windows, error budgets, multi-window burn alerts, and policy when budget is exhausted. Use when defining reliability targets or aligning eng and product on trade-offs.

Deep OAuth 2.0 / OpenID Connect workflow—choosing flows for client type, PKCE, tokens and validation, scopes and consent, rotation, and common misconfigurations. Use when implementing SSO, social login, or API access on behalf of users.

MirrorRoom 3.8 — Hybrid Bedrock Patch. The Mirror & The Room interactive research app. Conference-safe + downloadable. Runs two AI interaction modes: Flat (standard assistant) and Mirror (QSM-informed, with //inner thoughts// and recursive play). Now with optional proxy routing for hosted/demo environments, Rune seed, Audit the Judge feature, and richer seed ecology. Paste MirrorRoom-3.8.jsx into Claude.ai as a React artifact to run it.