Agi Agent Kit

Send emails through Resend's official MCP server. Supports single send, batch send, HTML and plain text bodies, attachments, CC/BCC, scheduling, and contact management.

Query Google NotebookLM notebooks from Claude Code for source-grounded, citation-backed answers. Browser automation, library management, persistent auth.

Extract text from PDF files for manipulation, search, and reference. Use when needing to read PDF content, extract text from documents, search within PDFs, or convert PDF to text for further processing. Supports multiple extraction methods (pdfplumber, PyMuPDF, pdfminer) with automatic fallback.

Documentation harvesting agent for crawling and extracting content from documentation websites. Use for crawling documentation sites and extracting all pages about a subject, building offline knowledge bases from online docs, harvesting API references, tutorials, or guides from documentation portals, creating structured markdown exports from multi-page documentation, and downloading and organizing technical docs for embedding or RAG pipelines. Supports recursive crawling with depth control, content filtering, and structured output.

Export current session context as a portable briefing document for Claude on claude.ai (Cowork). Use when the user wants to delegate tasks to Claude web, schedule work in Cowork, or hand off context to a remote Claude instance. Triggers on: 'export to cowork', 'cowork briefing', 'send to claude.ai', 'schedule in cowork', 'remote agent context'.

Automated documentation maintenance and generation skill. Triggers when: (1) Code is added, changed, updated, or deleted in any skill, (2) New scripts or references are created, (3) SKILL.md files are modified, (4) User requests documentation updates, (5) Skills catalog needs regeneration, (6) README or AGENTS.md need updates reflecting code changes. Use for generating technical documentation, updating docs after code changes, producing changelogs, ensuring documentation stays synchronized with the codebase, and maintaining the skills catalog.

Intelligent token optimization through Qdrant-powered semantic caching and long-term memory. Use for (1) Semantic Cache - avoid LLM calls entirely for semantically similar queries with 100% token savings, (2) Long-Term Memory - retrieve only relevant context chunks instead of full conversation history with 80-95% context reduction, (3) Hybrid Search - combine vector similarity with keyword filtering for technical queries, (4) Memory Management - store and retrieve conversation memories, decisions, and code patterns with metadata filtering. Triggers when needing to cache responses, remember past interactions, optimize context windows, or implement RAG patterns.

Synchronize upstream fork repositories with the agi-agent-kit codebase. Detects new/updated content from forked repos, creates feature branches, and applies framework adaptations.

Platform-adaptive plugin and extension auto-discovery. Detects the runtime environment (Claude Code, Gemini, Opencode, Kiro) and recommends or installs relevant plugins, extensions, MCP servers, and marketplace integrations. Use when setting up a project, onboarding, or when the user asks about available tools/plugins.

Answer questions about Claude Code / Anthropic agent design patterns (CLAUDE.md layering, skills anatomy, progressive disclosure, memory-first, orchestration, sub-agents, worktree isolation, Karpathy loop). Primary backend is a NotebookLM RAG over Anthropic design notebooks; falls back to bundled references when the notebook is not registered.

Generates images via Openrouter API using AI image models. Supports two modes: test (cheap model for iteration) and production (high-quality model for final output). Handles prompt construction, API calls, base64 decoding, and file saving. Supports reference images (logos, mascots) for brand-consistent generation.

INTERNAL ONLY. Forked write-capable variant of the public notebooklm skill. Adds programmatic source ingestion (add_source) for the X/YouTube → NotebookLM pipeline (#119). Headless, non-interactive, cron-driven. NEVER ship publicly.

Supply chain threat intelligence monitor. Scrapes TheHackerNews.com for supply chain security incidents and extracts compromised package names. Triggers on: 'check supply chain', 'supply chain threats', 'blocked packages', 'scan for compromised packages', 'vet upstream skills', 'refresh blocklist', before accepting upstream skills, before releases.