Loading...
Loading...
Found 105 Skills
Hand off to a fresh Claude session. Use when context is full, you've finished a logical chunk of work, or need a fresh perspective. Work continues from hook.
Integrate Honcho memory and social cognition into existing Python or TypeScript codebases. Use when adding Honcho SDK, setting up peers, configuring sessions, or implementing the dialectic chat endpoint for AI agents.
Retrieval-Augmented Generation patterns for grounded LLM responses. Use when building RAG pipelines, constructing context from retrieved documents, adding citations, or implementing hybrid search.
Manage long-running agent sessions. Use for tracking progress in extended tasks, maintaining context across long sessions, and managing multi-step workflows.
Recursive Language Model context management for processing documents exceeding context window limits. Enables Claude to match Gemini's 2M token context capability through chunking, sub-LLM delegation, and synthesis.
Context management in GPUI including App, Window, and AsyncApp. Use when working with contexts, entity updates, or window operations. Different context types provide different capabilities for UI rendering, entity management, and async operations.
Comprehensive knowledge of Claude Agent SDK architecture, tools, hooks, skills, and production patterns. Auto-activates for agent building, SDK integration, tool design, and MCP server tasks.
Based on the Recursive Language Models (RLM) research by Zhang, Kraska, and Khattab (2025), this skill provides strategies for handling tasks that exceed comfortable context limits through programmatic decomposition and recursive self-invocation. Triggers on phrases like "analyze all files", "process this large document", "aggregate information from", "search across the codebase", or tasks involving 10+ files or 50k+ tokens.
Agent Orchestration Rules
Maintain durable project context in `tasks/context.md` (state, decisions, milestones, gotchas, optional context links), inline during other workflows or standalone for cleanup/backfill. Triggers: update context.md, decision log, record project context, capture high-value reference links that improve context handoff.
Progressive context refinement pattern for subagents. Solves the problem of agents not knowing what context they need until they start working. Uses a 4-phase loop: DISPATCH, EVALUATE, REFINE, LOOP.
Use when user asks 'how should I...' or 'what's the best approach...' after exploring code, OR when you've tried to solve something and are stuck, OR for unfamiliar workflows, OR when user references past work. Searches conversation history.