prompt-pipeline
How to design the numbered prompt pipeline that drives Hunt phases in Cavekit. Covers greenfield 3-prompt patterns, rewrite 6-9 prompt patterns, shared…
Skills
Find an Agent Skill for your task. Read the guide and copy the install command.
Updated 11-03 00:11
Agents & models
25,539 results
validation-first
Validation-first design for AI agent output — every spec requirement must be automatically verifiable. Covers the 6-gate validation pipeline, phase gates…
livestreaming to atprotocol
Guide for publishing agent activity, responses, and reasoning to ATProtocol in real-time via Letta Code hooks. Use when setting up transparent "glass box" AI…
codex
When Claude or another primary assistant receives a user message, by default it hands it to Codex via the local shared Codex App Server for processing, including questions, discussions, and task execution, without the user needing to name Codex. The primary assistant is responsible for dispatching, relaying, and verifying results; the user chooses continuous tracking or dispatch-only, and the phone can access the same session via ChatGPT SSH.
hackathon-judging
Steers Kaggle Hackathon hosts toward defensible LLM-assisted grading workflows collecting writeups and artifacts via the Kaggle MCP server, auditing with…
ai-interaction
Continuous communication channel via MCP AI Interaction tool. Activate with 'khởi động ai_interaction'. Enables real-time Vietnamese conversation with…
jujutsu-colocated
Safe, practical workflows for using Jujutsu (`jj`) in colocated repositories (`.git` + `.jj`) and parallel-agent environments. Use when planning, editing,…
skulk
Wait for something without sleep-polling: schedule an OpenCode session wake-up with skulk. Use whenever an agent needs to wait for CI, GitHub Actions runs,…
llm-annotation-guide
Use this skill when a developer wants to annotate their LLM outputs, set up an annotation process, or improve existing annotations. Triggers on: "help me…
llm-judge-creator
Use this skill when a developer wants to create LLM-as-a-judge evaluators for their AI system. Triggers on: "create an LLM judge", "build an eval for my AI",…
git-cli-agentic
Git commands with porcelain and machine-readable output for agent workflows. Use when scripting status/diff/log, porcelain=v2, --numstat counts, custom…
trae-skill-finder
Discover and install skills for Trae IDE. Use when user asks 'find a skill', 'is there a skill for X', 'how do I do X'. Wraps find-skills with Trae-specific…
developer-scrum-master
The Developer Scrum Master facilitates Agile methods, manages the sprint backlog, designs sprints, translates the PRD into developer tasks (tickets), and…
skill-creator
Guide for creating effective skills. This skill should be used when users want to create a new skill (or update an existing skill) that extends Claude's…
content-strategy
The English introduction is being prepared.
run-gemini
Send an assistance request to the assistant, suitable for task scenarios such as long-text reading comprehension, web search, web page reading, article translation, and web and front-end interface creation.
mcp-oauth
Add OAuth 2.0 PKCE authentication to a remote MCP server. Use this skill whenever the user wants to add authentication to an MCP server, protect MCP tools with…
search
Search the web, library documentation, and GitHub repositories using Tavily, Context7, GitHub Grep, Exa fallback, and Z.AI Web Search Prime MCP fallback. Use…
reflect
Reflect on the current session and suggest AGENTS.md improvements based on user corrections, repeated feedback, and patterns observed during interaction. Use…
agent-browser
Browser automation CLI for AI agents. Use when the user needs to interact with websites, including navigating pages, filling out forms, clicking buttons, taking screenshots, extracting data, testing web apps, or automating any browser task. Triggers include "open a website", "fill out a form", "click a button", "take a screenshot", "scrape data from a page", "test this web app", "log in to a site", "automate browser actions", or any task requiring programmatic web interaction.
swift-lang
Swift 6.2 language patterns and critical corrections for LLM code generation. Covers the concurrency paradigm shift (default MainActor isolation, @concurrent…
context7
Use this skill when you need to work with context7 through its generated async Python app, call its MCP-backed functions from code, or inspect available…
claude-code-mastery
The definitive Claude Code setup, configuration, and mastery skill. Use when setting up Claude Code for a new project, optimizing an existing setup,…
critic
Challenge assumptions and probe reasoning using adversarial thinking. Use when asked to find weaknesses, challenge a design, identify edge cases, or…