Loading...
Loading...
Found 37 Skills
Reference guide for Agentica multi-agent infrastructure APIs
Classify user requests and route to the correct agent + skill combination. Use for any user request that needs delegation: code changes, debugging, reviews, content creation, research, or multi-step workflows. Invoked as the primary entry point via "/do [request]". Do NOT handle code changes directly - always route to a domain agent. Do NOT skip routing for anything beyond pure fact lookups or single read commands.
Research collection of reconstructed prompt patterns and architectures for agentic AI coding assistants
Set up and improve harness engineering (AGENTS.md, docs/, lint rules, eval systems, project-level prompt engineering) for AI-agent-friendly codebases. Triggers on: new/empty project setup for AI agents, AGENTS.md or CLAUDE.md creation, harness engineering questions, making agents work better on a codebase. ALSO triggers when users are frustrated or complaining about agent quality — e.g. 'the agent keeps ignoring conventions', 'it never follows instructions', 'why does it keep doing X', 'the agent is broken' — because poor agent output almost always signals harness gaps, not model problems. Covers: context engineering, architectural constraints, multi-agent coordination, evaluation, long-running agent harness, and diagnosis of agent quality issues.
AI agents as force multipliers for quality work. Core skill for all 19 QE agents using PACT principles.
Use when the user needs to build AI agents — tool use patterns, memory management, planning strategies, multi-agent coordination, evaluation, and safety guardrails. Triggers: user says "agent", "build an agent", "tool use", "agent loop", "multi-agent", "memory management", "guardrails", "agent evaluation".
N coordinated agents on shared task list using tmux-based orchestration
This skill should be used when the user asks to "implement a feature in an isolated worktree", "create a worktree from the current project branch", "open a PR from worktree changes", "merge feature PRs into main", "run multiple agents in parallel worktrees", or "handle worktree merge conflicts and incompatibilities".
Persistent cross-session task queue for AI agents using Claude Code Tasks schema. Add, claim, complete, and reassign tasks with move-based locking, dependency tracking (blocks/blockedBy), conversation transcript linking, and staleness detection. Use for: (1) saving tasks for future agent sessions, (2) cross-session task persistence, (3) multi-agent task coordination, (4) linking conversation transcripts to tasks. Triggers: task queue, save task, agent task, queue task, persistent task, cross-session task, task for agent.
Route issue-running automation through a deterministic control plane that selects agent + model from registry, can coordinate multiple safe parallel agents, and executes the unified run-agent runner.
Multi-agent coordination discipline: one-message-then-wait (send complete context, wait for reply before sending again), idle notifications are heartbeats (no action unless extended + blocking + user asked), no polling loops (event-driven only), never fabricate agent responses (wait for real system events), sequential agent spawning (acknowledge between each), and proper shutdown protocol (request, wait, respect rejection). Activate when orchestrating multiple agents, managing agent teams, coordinating handoffs between agents, spawning subagents, or building multi-agent workflows. Triggers on: "coordinate agents", "spawn multiple agents", "manage agent team", "agent keeps sending messages", "polling loop", "agent idle", "shut down agent", "multi-agent workflow", "agent handoff", "coordinate parallel work", "stop bothering the other agent". Also relevant when an agent is fabricating responses, sending follow-up messages before replies arrive, or reacting to idle notifications unnecessarily.
Manus-style context engineering for Agent Teams. Coordinate multiple Claude Code instances with shared planning files. Use when complex tasks need parallel work (code review, debugging, feature development). Requires CLAUDE_CODE_EXPERIMENTAL_AGENT_TEAMS=1.