toolfs
ToolFS は、LLM エージェントに統一された仮想ファイルシステムアクセスを提供するフレームワーク
Killer-Skillsディレクトリで数千のAI Agentスキルを探索・インストール。Claude Code、Windsurf、Cursorなどに対応。
ToolFS は、LLM エージェントに統一された仮想ファイルシステムアクセスを提供するフレームワーク
Create single-file commands with dynamic content injection (@path and !command). Use when building commands that need filesystem access, git state, runtime context, or argument handling with $1 placeholders. Includes @file injection, !shell execution, <injected_content> wrappers, and single-file structure patterns. Not for skills (use skill-development) or audit workflows.
This skill should be used when the user asks about plugin settings, store plugin configuration, user-configurable plugin, .local.md files, plugin state files, read YAML frontmatter, per-project plugin settings, or wants to make plugin behavior configurable. Documents the .claude/plugin-name.local.md pattern for storing plugin-specific configuration with YAML frontmatter and markdown content.
Preserve existing Markdown document writing style while editing. Use when modifying existing .md files to add new analysis results, sections, or notes without rewriting prior content. Prioritize append/insert over replace unless the user explicitly requests a rewrite; keep headings, terminology, table formats, and link style consistent with the existing document.
Pattern detection and automatic skill recommendation system. Activates when analyzing Cortex memory files, detecting recurring work patterns, or determining if new skills are needed. Analyzes .cortex_log.md, PRD files, and task lists to identify patterns (API calls, testing, deployment, etc.) appearing 5+ times. Generates Synapse_RECOMMENDATIONS.md with prioritized skill suggestions. Use when optimizing workflows or identifying automation opportunities.
Triage and process WhatsApp action files from the AI Employee vault. This skill should be used when WHATSAPP_*.md files appear in /Needs_Action folder, or when user asks to handle WhatsApp messages, draft replies, or generate invoices from chat requests. Reads message content, detects intent, creates Plan.md with steps, and always routes replies to /Pending_Approval — never sends directly to WhatsApp without human approval.
Build CEFR-split word tree markdown files from the dated daily vocab note in the current language folder. Use this when the user wants to read `vocab/daily/daily-notes-YYYY-MM-DD.md`, organize every vocab item by part of speech, expand each item with usable grammar details, verify CEFR level through web search, rewrite the daily vocab note into the normalized word-tree schema, write dated `vocab-CEFR-YYYY-MM-DD.md` files inside per-level `daily/` folders, and update the cumulative CEFR vocab files.