gauntlet-loop
Domain-agnostic method for producing top-tier work with the Gauntlet Loop — split → build → blind-critic → repeat — against a hard "bar" the agent cannot argue…
Skills
Find an Agent Skill for your task. Read the guide and copy the install command.
Updated 11-05 00:16
Agents & models
25,653 results
memory-integration
Use to maintain context across sessions - integrates episodic-memory for conversation recall and mcp__memory knowledge graph for persistent facts
autoresearch-agent
Delegate autoresearch tasks to a headless Claude Code sub-agent process. Use when the current agent needs to run any autoresearch command (autoresearch, ship,…
agent-instructions
Create or reconcile scoped AGENTS.md, CLAUDE.md, and Cursor rules from repository evidence. Use when coding-agent guidance is missing, stale, duplicated, or…
skill-creator
Create, improve, merge, and validate portable Agent Skill packages with adaptive workflows and behavioral evals. Use when authoring or reconstructing SKILL.md…
velt-chat-sdk-adapter-best-practices
Velt Chat SDK Adapter implementation patterns and best practices for building bots that integrate with Velt comment threads. Use when building a bot (greeting…
allocate-agent-models
Use when delegating OpenAI agent work across GPT-5.6 Sol, Terra, or Luna, especially for new task classes, parallel workers, independent review, model…
factory-setup
Connect a third-party coding agent (Claude Code, Codex, Cursor, or any other MCP-capable client) to Warp Factory by installing and authenticating the oz CLI…
research-hub-multi-ai
Research-domain router that writes `.coord/multi_ai_plan.md` when one round needs two or more bounded delegates and touches `.research/`, `.paper/`, Zotero,…
agent-right-brain
Give agents creative abilities using `rawgenai` — speak, listen, generate images/videos/music/sound effects, create multi-speaker dialogue, and manage voices.…
deepseek-harness-plugin-creator
Scaffold, create, materially revise, validate, or debug external DeepSeek Harness/Cordis plugins and bundle layers, including Tools, Services, schedulers,…
readme-agents-writer
Build and maintain unified README.md + AGENTS.md documentation with a canonical README and an AGENTS.md symlink. Use for writing or refactoring verified human…
mcp-builder
Guide for creating high-quality MCP (Model Context Protocol) servers that enable LLMs to interact with external services through well-designed tools. Use when…
xberg-chunking
Use when splitting extracted text into chunks for LLM context windows or RAG ingestion. Covers chunk size, overlap, markdown/yaml/semantic chunkers,…
xberg-extracting-keywords
Use when extracting keywords (YAKE/RAKE) from documents — and, secondarily, when detecting document language or generating embeddings for RAG and search.…
rock-agent-debug
Troubleshoot Agent Jobs running in the ROCK sandbox. Supports two types: Harbor Job and Bash Job. When the user provides a sandbox ID and asks about job status, progress, logs, reward, agent execution, or encounters job failure, trial anomalies, reward of 0, agent...
ilya-sutskever-perspective
Ilya Sutskever's thinking framework and way of expression. Based on deep research of 12 first-hand conversations, 9 academic papers, 10 hours of sworn testimony, 27 recommended reading lists, and 14 authoritative secondary sources, distilling 6 core mental models, 8 decision heuristics, and a complete expression DNA. Purpose: as a thinking advisor, to analyze AI technology directions, safety strategies, and research taste from Ilya's perspective. When users mention "from Ilya's perspective"...
codex-orchestrator
DEFAULT PIPELINE for all tasks requiring execution. You (Claude) are the strategic orchestrator. Codex agents are your implementation army - hyper-focused…
run-codex-subagents
Use skill if you are orchestrating Codex work through the released codex-worker CLI and need correct run/send/thread/turn, request-response, wait, logs, or…
transformers
Work with state-of-the-art machine learning models for NLP, computer vision, audio, and multimodal tasks using HuggingFace Transformers. This skill should be…
claude-hud
Configure Claude Code statusline with context window monitoring using CC 2.1.6 fields. Use when configuring statusline, monitoring context, displaying HUD.
advanced-evaluation
Implementation of "LLM-as-a-Judge", comparison of outputs, creation of evaluation rubrics and bias mitigation in AI systems.
multi-agent-patterns
Multi-agent system architectures, orchestration patterns, swarms, handoffs, and consensus protocols to overcome single-context limits.
project-development
Methodology for developing agent-based projects, from ideation and validation of "Task-Model Fit" to the architecture of pipelines of…