智能体
Agent 规划、多步流程与编排
ask-matt
CartmanFatass
Ask which skill or flow fits your situation. A router over the skills in this repo.
adversarial
Mike-Byrne-AI
Failure-mode discovery — the ways future-me forgets a step, an AI collaborator drifts, or a refactor misses a sister-site. Use when shipping a hook, gate, or governance change; before a release; when you want a pre-mortem before the implementation lands. Distinct from /review (per-file correctness) and /reflect (cross-artifact gap surfacing). Delegates to the failure-mode-reviewer agent.
rapid
pxdogbo
Rapid session — capture realtime notes from the user while working with a live product (app, website, etc.) without losing context. Bare /rapid (or /rapid start with no slug) opens an instant menu — new session, resume, cleanup, review, handoff, update, help — and does nothing until you pick, so starting never blocks on housekeeping or the network. /rapid <note> skips the menu and captures straight into a new-or-reused session (fresh doc + sibling git worktree on a rapid/<slug> branch). Cleanup of finished sessions only ever runs when you ask for it (the menu's Cleanup option, or tidy/burn) — never automatically at start. Drive-by notes and bare-word triggers (review/recap, push (→ always commits the queue and opens a PR, never a bare git push), carpool, wash/clean, park, unpark, drop, test/testdrive, scrap, tidy, burn, link, reverse/undo, inbox) operate on the session this chat started, not on whatever doc happens to be marked active globally. wax grooms the doc in place (condense + group + de-stale) without emptying the queue. handoff seeds a scoped plan as a standalone session a fresh chat adopts; collab is a chatroom between two live agents; inbox leaves an async note in another session's doc (no loop, no poke — it never triggers the peer; to task or talk to a live agent use collab) for it to pick up later. Use whenever the user types /rapid, /rapid <note>, /rapid review, /rapid done, /rapid off, /rapid update, /rapid handoff, or says they want to "start a rapid session" / "drop a quick note" mid-task.
hmasd-chatgpt-pro-transport
CartmanFatass
In the current authoring session, send one committed HMASD Pro question through Agentify, arm deterministic observation, read the complete answer at the assigned repository target or preserve a complete chat fallback, and close the tab. Never science, never a resend.
OpenClaw ACI Bridge
agerelated-clioquinol357
Agent-Computer Interface for autonomously operating Windows desktop applications and web browsers via unified JSON commands. Supports UIA, cursor probing, OCR, contour detection, vision VLM, and persistent pseudo-UIA knowledge.
orchestrate-sessions
Munyon-Canyon
Runs several Claude sessions on one machine as a team that lands milestone PRs on staging all day. One main orchestrator supervises, one dispatcher turns tickets into stacks, one CI manager keeps local stage 0 and CI fast so agents get PRs up with the least wait. Use when asked to be the main orchestrator over other sessions, to be the CI manager or CI optimization agent, to keep sessions unblocked and PRs merging, to maximize PRs landed, or to run the milestones unattended while the user is away.
moai-workflow-testing
modu-ai
Comprehensive testing and development workflow specialist combining DDD testing, characterization tests, performance profiling, code review, and quality assurance. Use when writing tests, measuring coverage, creating characterization tests, performing TDD, running CI/CD quality checks, or reviewing pull requests. Do NOT use for debugging runtime errors (use expert-debug agent instead) or code refactoring (use moai-workflow-ddd instead).
moai-foundation-context
modu-ai
Manages context window optimization, session state persistence, and token budget allocation for multi-agent workflows. Use when dealing with token budget management, context window limits, session handoff, state persistence across agents, or /clear strategies. Do NOT use for agent orchestration patterns (use moai-foundation-core instead).
01-readonly-inventory
tin-computer
Extract stable project knowledge without altering or overstating source material.
create-workflow
tin-computer
Author a bounded Tin workflow package and reviewable qualification cases from a brief.
product-analytics
tin-computer
Read one connected PostHog project and deliver a recurring, evidence-based product analytics brief.
Kiln
codeofaxel
Docs
moai-workflow-tdd
modu-ai
Test-Driven Development workflow specialist using RED-GREEN-REFACTOR cycle for test-first software development. Use when developing new features from scratch, creating isolated modules, or when behavior specification drives implementation. Do NOT use for refactoring existing code (use moai-workflow-ddd instead) or when behavior preservation is the primary goal.
scenario-blender-texturing-shading
scenario-labs
"Use when texturing, shading or look-developing in Blender: building a PBR material from a texture set, displacement, procedural materials, edge wear, dirt, grime, color variation, decals, box or triplanar mapping on meshes without UVs, stylized character skin (subsurface, pores), eyes and cornea, hand-painted stylized textures, texture painting or painting masks headless. Also when a material looks plastic, CG, waxy or blown out, color spaces are wrong, or EEVEE and Cycles disagree."
scenario-maya-animation
scenario-labs
"Use when animating a character or prop in Maya through Python: a walk or run cycle, a jump or landing, a heavy lift, throw or punch, an acting or dialogue shot, lip sync, eye darts and blinks, stepped blocking, splining, polish, Graph Editor tangents, animation layers, mocap cleanup or retargeting, game clips and root motion, or reviewing a playblast. Also when motion looks floaty, weightless, mushy, poppy, posey, sliding or chattery, or feet slap or skate."
context-fundamentals
guanyang
This skill should be used when the user asks to "understand context", "explain context windows", "design agent architecture", "debug context issues", "optimize context usage", or discusses context components, attention mechanics, progressive disclosure, or context budgeting. Provides foundational understanding of context engineering for AI agent systems.
opencode
NousResearch
Delegate coding tasks to OpenCode CLI agent for feature implementation, refactoring, PR review, and long-running autonomous sessions. Requires the opencode CLI installed and authenticated.
octocode-roast
bgauryy
This skill should be used when the user asks to "roast my code", "review code brutally", "find code sins", "what's wrong with my code", "shame my code", "critique this code", "find antipatterns", "code quality roast", or wants entertaining but actionable code criticism with severity-ranked fixes. Delivers brutally honest roasts with file:line citations and redemption paths.
octopus-quick-review
nyldn
Fast pre-commit code review using multi-AI consensus. Use when: Use this skill when the user says "review this code", "check this PR",. "quality check the implementation", "review my changes", or "what's wrong with this code".
octopus-quick
nyldn
Quick execution for ad-hoc tasks without workflow overhead. Use when: Use this skill when user says "quick fix", "ad-hoc task", or explicitly. requests fast execution without full workflow overhead.
flow-next-epic-review
gmickel
Epic completion review - verifies all epic tasks implement spec requirements. Triggers on /flow-next:epic-review.
context-engineering-collection
muratcankoylan
A comprehensive collection of Agent Skills for context engineering, multi-agent architectures, and production agent systems. Use when building, optimizing, or debugging agent systems that require effective context management.
create-hat-collection
mikeyobrien
Generates new Ralph hat collection presets through guided conversation. Asks clarifying questions, validates against schema constraints, and outputs production-ready YAML files.
git-workflow
agno-agi
Git workflow guidance for commits, branches, and pull requests