Loading...
Loading...
Found 4,459 Skills
Validates Claude Code plugins against architectural best practices for Agents, Skills, MCP, and Progressive Disclosure. Use when validating plugin structure, reviewing manifest files, checking frontmatter compliance, or verifying tool invocation patterns.
Spawn and manage multiple Codex CLI agents via tmux to work on tasks in parallel. Use whenever a task can be decomposed into independent subtasks (e.g. batch triage, parallel fixes, multi-file refactors). When codex and tmux are available, prefer this over the built-in Task tool for parallelism.
This skill should be used when the user asks to "optimize with SIMBA", "use Bayesian optimization", "optimize agents with custom feedback", mentions "SIMBA optimizer", "mini-batch optimization", "statistical optimization", "lightweight optimizer", or needs an alternative to MIPROv2/GEPA for programs with rich feedback signals.
Discover, create, and validate headless adapters for agent integration. Includes scaffolding tools and schema-driven compliance testing.
An AI Agent Skill that enforces a 'Risk Triage -> Align -> Act' protocol. Triggers when requests are vague, conflict-ridden, or high-impact. Do not activate for low-risk or precise, scoped requests.
Creates Cursor-specific AI subagents with isolated context for complex multi-step workflows. Use when creating subagents for Cursor editor specifically, following Cursor's patterns and directories (.cursor/agents/). Triggers on "cursor subagent", "cursor agent".
Test, validate, and improve agent instructions (CLAUDE.md, system prompts) using sub-agents as experiment subjects. Measures instruction compliance, context decay, and constraint strength. Use for "test prompt", "validate instructions", "prompt effectiveness", "instruction decay", or when designing robust agent behaviors.
Design effective system prompts for custom agents. Use when creating agent system prompts, defining agent identity and rules, or designing high-impact prompts that shape agent behavior.
Recovery protocols when agent is stuck—escalate to new agent, migrate context to new session, or reset mid-conversation.
Automatically fix ESLint errors by modifying code to comply with linting rules. For small codebases (≤20 errors), fixes directly. For larger codebases (>20 errors), spawns parallel agents per directory for efficient processing. Never disables rules or adds ignore comments.
Generate a plan for how an agent should accomplish a complex coding task. Use when a user asks for a plan, and optionally when they want to save, find, read, update, or delete plan files in $CODEX_HOME/plans (default ~/.codex/plans).
Agent Orchestration Rules