Skills
97,725 packaged capabilities and playbooks for agents. Install any of them with one command.
Agent workflows
Use when you need to locate the current spec pack (FEATURE_DIR) in the sdlc-dev Spec workflow, avoid reading or writing requirements/*.md in the wrong directory, or when issues arise such as "wrong context / wrong file written / branch not following the convention".
Look up official SurrealDB Agent Memory (formerly Spectron) documentation from surrealdb.com as markdown: quickstarts, mental model, ingest, sessions, recall, reasoning, SDKs, framework integrations, MCP server, REST API, CLI, and configuration. Use when building with or answering questions about SurrealDB Agent Memory. Read-only; sends only search terms and doc paths.
Session management and project status dashboard. Owns the full session lifecycle (start/work/close/resume), focus lane, bookmarks, worktree detection, and tab naming. Also serves as the project status dashboard — shows active epics, progress, actionable next steps, blocked items, tasks, GitHub issues, and recommendations. Worktree creation is deferred to swain-do task dispatch (SPEC-195). Triggers on: 'session', 'status', 'what's next', 'dashboard', 'overview', 'where are we', 'what should I work on', 'show me priorities', 'bookmark', 'focus on', 'session info'.
Use when the user wants to view AI conversation token usage as a receipt, invoice, checkout slip, token bill, usage receipt, cost snapshot, or creative monospace thermal-paper artifact. Always consider this skill for Chinese prompts like 查看本次对话 Token 消耗, 生成 token 小票, 对话发票, AI 用量账单, 把这次对话打成小票, 看看这轮 token 消耗, or any request to make token/context usage visually shareable.
Create a new AI Skill in an existing mini program project. Triggered when the user wants to add custom AI capabilities to an existing project (rather than installing a ready-made community Skill). Full flow: understand requirements → design interfaces → call wxa-skills-generate to generate code → pass validation with wxa-skills-validate.
A focused development workflow engine for AI coding agents, invoked as `/absolute <command> [target]`. One skill, eleven commands. One sets it up: `init` (interview how you want absolute to behave + detect the stack once, then write JSON config — `.absolute.config.json` per project and `~/.absolute/config.json` for user defaults and per-project overrides — that every other command reads instead of re-detecting; non-blocking, the others soft-recommend it). Five cover the build loop you run every day — think → spec → plan → build → polish → document: `work` (end-to-end, phase-gated SDLC: relentless design interview → reviewed spec → dependency-graphed task board → safe-wave TDD execution → verification), `spec` (lightweight standalone design spec: codebase scan → bounded clarify pass → write a reviewed design doc to docs/plans → scored reviewer subagent → stop, no build), `ui` (polished, intentional interface design with concrete CSS/Tailwind values — typography, color, layout, dark mode, accessibility, animations), `simplify` (autonomously simplify your working git changes — reduce complexity, remove redundancy, keep tests green), and `docs` (Diátaxis-driven documentation — tutorials, how-tos, reference, explanation, READMEs, ADRs, with stack detection and code-verified accuracy). Five more keep the codebase healthy — standing maintenance on green main, each a risk-ranked, gated, safe-wave scan-and-fix: `upgrade` (dependency upgrades — outdated/vulnerable deps bumped in semver waves, majors gated and changelog-read, tests green after each), `audit` (vulnerability & security scan — dependency CVEs plus risky code patterns, severity-triaged and remediated; defensive, on your own repo), `prune` (dead code & dependency cleanup — unused deps, exports, files removed with evidence, repo-wide), `debt` (lint & typecheck debt paydown — clear pre-existing warnings/errors and suppressions by rule, no cheating the checker), and `deflake` (flaky test fixes — diagnose and fix the root nondeterminism, never retry/skip/sleep). Triggers on "absolute init|work|spec|ui|simplify|docs|upgrade|audit|prune|debt|deflake", "set up absolute", "initialize absolute", "configure absolute", "build this end-to-end", "plan and build", "grill me on this", "write a spec", "draft a design doc", "design this UI", "make this less like AI slop", "simplify this", "clean up my changes", "write docs", "audit our docs", "upgrade our dependencies", "are we vulnerable", "security audit", "remove dead code", "fix our lint/type errors", "fix flaky tests", or any multi-step development, design, cleanup, documentation, or maintenance task.
Archive an OpenSpec change end-to-end. This skill invokes openspec-archive-change or openspec-bulk-archive-change itself to perform the native merge, then immediately follows up with the cross-capability linking and running indices OpenSpec doesn't build on its own — linking every capability a change touched via a shared `related:` frontmatter list, keeping `openspec/specs/INDEX.md` current, and appending a row to `openspec/changes/archive/INDEX.md`. Use this skill whenever the user wants to archive a change, says "archive this change", "bulk archive these changes", "run openspec-archive-change", "run openspec-bulk-archive-change", "update the specs index", "cross-link the specs", or wants the archived-change changelog kept current. This skill is purely additive on the linking side — it never prunes a `related:` entry and never changes a change's `Status` column after the initial write; that pruning/synthesis work belongs to `accelint-archive-synthesis`.
Multi-agent design patterns for Claude Code and AI development systems. Covers custom subagents, built-in subagents (Explore/Plan/General-purpose), fungible vs specialized agents, delegation, orchestration, batch workflows, context hygiene, and failure recovery. Use when: designing multi-agent systems, creating custom subagents, delegating bulk operations, orchestrating parallel work, choosing between fungible and specialized agents, configuring agent frontmatter, or debugging agent coordination issues.
Patterns and techniques for evaluating and improving AI agent outputs. Use this skill when implementing self-critique and reflection loops; building evaluator-optimizer pipelines for quality-critical generation; creating test-driven code refinement workflows; designing rubric-based or LLM-as-judge evaluation systems; adding iterative improvement to agent outputs (code, reports, analysis); or measuring and improving agent response quality.
Autonomously optimize any Claude Code skill by running it repeatedly, scoring outputs against binary evals, mutating the prompt, and keeping improvements. Based on Karpathy's autoresearch methodology. Use when: optimize this skill, improve this skill, run autoresearch on, make this skill better, self-improve skill, benchmark skill, eval my skill, run evals on. Outputs: an improved SKILL.md, a results log, and a changelog of every mutation tried.
Monitors awesome-copilot releases for drift against the amplihack integration. Checks latest commits on github/awesome-copilot via the GitHub API and reports whether the local integration is current or has drifted behind upstream changes. Use when auditing integration freshness or before updating awesome-copilot features.
How to write and maintain DESIGN.md as the visual specification layer for Cavekit projects. Nine-section format, design tokens, accessibility, integration with kits/plans. Use when defining or revising visual identity, importing a third-party design system, or auditing UI code against design tokens.
Validation-first design for Cavekit — every kit requirement must be automatically verifiable. Six-gate validation pipeline, phase gates, merge protocol, completion signals, acceptance criteria patterns. Use when designing acceptance criteria, wiring CI gates, or debugging why an agent is producing output that nobody can verify.
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.
Builds and debugs Letta Code channels, including first-party channel adapters and dynamic user channel plugins under ~/.letta/channels. Use when adding Telegram, WhatsApp, Bluesky, Slack, Discord, or custom channel support; testing channel routing, pairing, MessageChannel, runtime dependencies, or channel plugin manifests.
Set up a full AI ensemble/mob programming team for any software project. Creates agent definition files (.claude/agents/), coordinator instructions (.team/coordinator-instructions.md), project owner constraints (PROJECT.md), team conventions (AGENTS.md), architectural decisions (docs/ARCHITECTURE.md), domain glossary, and supporting docs. Use when: (1) starting a new project and wanting a full expert agent team, (2) the user asks to "set up a team", "create a mob team", "set up ensemble programming", or "create agent profiles", (3) converting an existing project to the driver-reviewer mob model, (4) the user wants AI agents to work as a coordinated product team with retrospectives and consensus-based decisions.
One-click installer for gate-cli and gate-skills on Cursor, Claude Code, Codex, or OpenClaw/mcporter. Triggers on 'install gate-cli', 'Gate skills', 'Gate Verify', 'tx check-in', 'setup Gate Cursor', 'Gate Claude Code', 'Gate Codex', 'OpenClaw Gate', 'mcporter Gate'.
Build and evaluate low-latency spoken, multimodal, and translation experiences with Google Gemini Live API and Gemini Enterprise Agent Platform Live API. Use when a media-production or voice-agent workflow needs real-time audio input/output, barge-in, voice configuration, live transcription, live translation, tool/function calling during a spoken session, WebSocket session design, latency QA, quotas/cost review, or Google/Vertex data-governance tradeoffs.
Integrate Arcjet Guard into the Python Claude Agent SDK — wrap authored @tool with guard_tool, and use guard_hooks for UserPromptSubmit inbound plus PreToolUse on unwrapped built-ins / MCP. Use when asked to add Arcjet to claude-agent-sdk, rate limit those tools, screen inbound prompts, or block prompt injection / PII. This is Python claude-agent-sdk, not the JS adapter and not Claude Managed Agents hosted sessions.
Integrate Arcjet Guard into official CrewAI — register process-wide PRE_TOOL_CALL with register_arcjet_hooks, or wrap a standalone BaseTool you call yourself with guard_tool. Use when asked to add Arcjet to CrewAI, LiteAgent, crew-injected tools, rate limit those tools, screen inbound messages before kickoff, or block prompt injection / PII. Official crewai only; not an npm CrewAI port and not LangChain Crew wrappers.
Use when starting, executing, or finishing non-trivial implementation tasks where reusable constraints may matter. Recall relevant lessons before work, capture reusable corrections or discoveries during and after work, and keep project memory in `docs/lessons/`.
Converts MCP servers into on-demand skills to cut context window usage, classifying each tool by replacement strategy and generating the skill package. Triggers on: "convert MCP", "MCP to skill", "reduce context size", "too many tools", "tool token bloat", "MCP migration".
Design and operate bounded multi-agent workflows with task decomposition, dependency graphs, ownership, handoff contracts, shared-state controls, approvals, recovery, and synthesis. Use when a task contains genuinely independent workstreams, specialized roles, parallel research or implementation, reviewer-worker loops, or coordination problems that one agent should not execute sequentially.
Prevent 此地无银三百两式 residue: finalize artifacts without echoing rejected session-only alternatives into labels, metadata, commits, PRs, or handoffs. Use after corrections or discarded proposals. Not for ordinary deletion, deprecation, migration, or exclusions materially required for safety, accuracy, compatibility, audit, quotation, or requested comparison.
Scaffold a complete cave plugin bundle — generates .cave-plugin/plugin.json manifest and the standard directory structure (commands/, skills/, agents/, themes/, hooks/). Use when a user wants to create, publish, or package a cave plugin for the marketplace. Triggered by "create a plugin", "scaffold a plugin", "/plugin create", or "new cave plugin".
Двусторонняя сверка кода и графа — уборочный такт: всюду ли код соответствует графу, а граф — коду; конец каждой сущностной задачи. Триггеры: reconcile, «сверь код с графом», «приведи граф в порядок», «протки свою область», «отдай долги графа», does the graph match the code, tidy the graph. Также: задача, записанная крией; феномен-ярлык состояния; комментарий-портянка, пересказывающая граф. Distinct from weaving (ремонт графа изнутри), integrity (claim-аудит), reality-audit (поведенческие заявки). Композирует entry, writing, weaving. Нужны тулы iskron_*.
Reviews Claude configuration files for security, structure, and prompt engineering quality. Use when reviewing changes to CLAUDE.md, agents, prompts, commands, hooks, or settings. Routes each file type to a targeted review skill and returns classified findings. Flags settings.local.json appearing in a changeset, hardcoded secrets, malformed YAML, insecure agent tool access, and unsafe hook commands. Does not review SKILL.md files — plugin-dev:skill-reviewer owns those.
Captures learnings, errors, and corrections to enable continuous improvement. Use when: (1) A command or operation fails unexpectedly, (2) User corrects Claude ('No, that's wrong...', 'Actually...'), (3) User requests a capability that doesn't exist, (4) An external API or tool fails, (5) Claude realizes its knowledge is outdated or incorrect, (6) A better approach is discovered for a recurring task. Also review learnings before major tasks.
Self-reflection + Self-criticism + Self-learning + Self-organizing memory. Agent evaluates its own work, catches mistakes, and improves permanently. Use when (1) a command, tool, API, or operation fails; (2) the user corrects you or rejects your work; (3) you realize your knowledge is outdated or incorrect; (4) you discover a better approach; (5) the user explicitly installs or references the skill for the current task
Iteratively improve a Claude Code skill using the skill-reviewer agent until it meets quality standards. Use when improving a skill with multiple quality issues, iterating on a new skill until it meets standards, or automated fix-review cycles instead of manual editing.
This skill should be used when the user asks to "create a status line", "customize status line", "set up statusline", "configure Claude Code status bar", "install ccstatusline", "add project colors to status line", "show git branch in status", "display token usage", or mentions Peacock colors, powerline, or status line configuration.
Export this agent's data into a migration bundle for import elsewhere. Use when moving an agent off Starchild or backing up state (e.g. export my memory and tasks, create a migration code, hand off to a new agent). Building a bundle works anywhere. Uploading it to the Starchild migration relay is only available from external agent platforms — a bundle built inside Starchild can be kept as a local backup but not handed off through the relay.
Create and train AI learning plugins with AgentDB's 9 reinforcement learning algorithms. Includes Decision Transformer, Q-Learning, SARSA, Actor-Critic, and more. Use when building self-learning agents, implementing RL, or optimizing agent behavior through experience.
New to MCP?
From what the protocol is to installing your first server, these answers cover most first questions.
An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.