prompts

Compare installed plugin versions against GitHub to confirm skills and agents are current. Use at session start, for "is everything up to date", "are my plugins up to date", "are our plugins current", "check for plugin updates", "is anything stale", "am I on the latest version", "did that plugin update land", or when a skill behaves differently than documented or references features you do not recognize. Checks any plugin in the marketplace, not just core. Completes in under 100ms per plugin.

This skill should be used for human-facing prose — emails, docs, reports, posts, release notes, and commit messages — when the user asks to "humanize", "make this sound less AI", "de-AI this", "this sounds like ChatGPT", or "edit this". Preserve facts, evidence, citations, house style, and intended meaning while removing clustered AI-writing patterns.

This skill should be used when the user asks for "game UI", "app-to-game UI", "video game HUD", "controller navigation", "D-pad navigation", "TV app", "ten-foot interface", "game UI key bindings", "controller-first menu", "console-style menu", "TV remote focus", or "directional navigation".

This skill should be used when tracking API spend, analyzing billing data, monitoring service costs, running budget reports, or optimizing agent spend across Anthropic, Vercel, Railway, and other platforms. Invoke when asked about "cost tracking", "billing analysis", "spend report", "API costs", "usage monitoring", "budget vs actual", "cost optimization", or "which agents cost the most".

This skill should be used when the user asks to "check deployment health", "verify service connectivity", "validate environment variables before deploy", "run a pre-deploy smoke check", or "diagnose a Vercel, Railway, Redis, or Postgres connection issue". Runs deterministic shell scripts for infrastructure health checks and environment validation, returning structured JSON instead of burning context on inline bash logic.

This skill should be used when the user asks to "integrate Plaid", "connect a bank account", "sync bank transactions", "add Plaid Link to my app", or "implement financial data access". Integrates the Plaid API for bank account connections and transaction syncing in TypeScript/Bun applications.

Finds stale and resource-hungry processes, scores them by waste, and presents a cleanup report with friendly names. Use this skill when the user says 'what's eating my RAM', 'kill stale processes', 'clean up my machine', 'free up memory', 'my computer is slow', 'what's running', 'too many things open', or asks to find/kill background processes. Also use proactively when you notice sluggishness, process spawn failures, or many duplicate processes during normal work.

Wait for CI/CD pipelines to finish after a push, then act on the result. Use after git push or opening a PR, and for "wait for CI", "check if the build passes", "monitor the pipeline", "is CI green?", or any push-then-verify cycle that must block on results. Supports GitHub Actions, GitLab CI, and Vercel deployments.

This skill provides guidance for building MCP Apps, the official io.modelcontextprotocol/ui extension for interactive HTML interfaces inside MCP hosts. This skill should be used when the user asks to "create an MCP App", "add UI to an MCP tool", "build interactive MCP", "ui:// resource", "sandboxed iframe MCP", "interactive chat UI", "embed UI in chat", "MCP tool with interface", or needs capability negotiation, resource CSP, app-only tools, host lifecycle, or progressive text fallbacks.

Get an independent read-only second opinion at a commitment boundary, before substantive work on a hard task, when stuck or changing approach, or at a final review gate. Use for "consult the advisor", "get a second opinion", "ask codex", "ask Fable", or "ask a bigger model". The advisor returns guidance; the main session keeps execution and decision ownership.

Run the Claude Code harness on GPT-5.6 Sol through a local CLIProxyAPI proxy when Anthropic usage runs out, and diagnose that setup when it drifts. Use for "my Anthropic usage ran out", "keep working on another model", "run Claude Code on GPT-5.6 Sol", "set up claudex", "claudex isn't working", or "bill against my Codex subscription". macOS + Homebrew.

Route bounded code-writing volume from a capable main session to cheaper or specialized executors — native subagents, Grok workers, Codex workers — while planning, design intent, review, verification, and git stay in the main seat. Use for "dispatch to workers", "plan big execute small", "race worker lanes", "model arbitrage", or "spec and dispatch".

This skill should be used when the user says "deploy a team", "spin up agents to work on this", "use all our agents", "coordinate specialists", or wants to break a large task into parallel sub-tasks handled by multiple domain experts simultaneously. Orchestrates Claude Code's experimental agent team system using the full core specialist roster.

Coordinate native specialist agents, external implementation workers such as Grok, and an independent advisor such as Fable from a capable main session. Use for "orchestrate this", "use Grok workers", "use Fable as advisor", "Codex main with workers", "delegate implementation but keep control here", or cross-model workflows. Never replaces the user's current main model.

Design or harden a software factory: an agentic loop that iterates toward a goal with a verification gate, persistent state, and a stop condition. Use for "build a loop", "agentic loop", "self-iterating agent", "/loop", "/goal", "Ralph loop", "maker-checker", "agentic SDLC", "ADW", or when picking a verification gate or bounding a loop's blast radius.

Dispatch many subagents in coordinated waves with per-wave review. Use for "fan out agents", "wave dispatch", "batch agents", "generate N variations", or any fan-out beyond about five parallel subagents that needs batching, ordering, and result reconciliation.

Comprehensive audit skill for agents and skills across the plugin ecosystem. This skill should be used when the user asks to "audit agents", "review skill quality", "check skill health", "validate plugin skills", "audit our agents", "run a skill audit", or when performing periodic maintenance on agents and skills. Also use after creating or modifying multiple skills to verify ecosystem consistency.

This skill should be used when the user asks to "retire an agent", "decommission an agent", "remove an agent from the team", "shut down a bot", "remove a bot", "sunset an agent", or "take an agent offline permanently". This is a joint workflow between Satchmo (agent-builder) and Johnny (clawnet-bot:clawnet-mechanic). Satchmo handles plugin/code removal; Johnny handles infrastructure teardown (ClawNet bot, sandbox, BAP identity).

This skill should be used when the user asks to "create a new agent", "onboard a new agent", "add an agent to the team", "deploy a new bot", "register an agent with Paperclip", or "add this agent to the roster". Provides the complete end-to-end checklist for bringing a new agent onto the bOpen team — design, write, avatar, plugin, Paperclip registration, roster, and optional ClawNet bot deployment.

Run skill benchmarks and write evals that score a skill against a no-skill baseline. Use for "benchmark this skill", "run the skill evals", "write an eval for this skill", or "did this skill actually help?".

This skill should be used when the user wants to publish a package to npm, bump a version, release a new version, or mentions "npm publish", "bun publish", "version bump", or "release to npm". Handles version bumping, changelog updates, git push, npm publishing, and automatic token rotation via agent-browser when auth expires. Do not trigger for unrelated uses of "release" (e.g. GitHub releases, press releases).

Split a large Claude Code or Codex plugin into a small core plus optional modules. Use for "split this plugin", "my plugin is too big", "reduce startup context", "Codex is omitting my skills", "core plus optional packs", or "publish part of my plugin separately". Covers marketplace subdirectory sourcing, the dependency field's silent-load behaviour, Codex adapter regeneration, and symlink depth. Use agent-auditor to measure first and benchmark-skills to verify routing after.

This skill should be used when the user asks to "add plugin settings", "make a plugin configurable", "store per-project plugin configuration", "use settings.local.json", "create a plugin state file", "expose skill settings in Agent Master", or "add a skill interface". Distinguishes official Claude Code settings from project-owned configuration and documents bOpen Agent Master skill interface discovery.

Use this skill when preparing to publish a package, plugin, or skill and you need human approval first. Invoked when the user says "publish", "release", "ship", or "push to registry" but no approved Linear ticket exists yet. Runs preflight checks, creates or updates a Linear ticket with a structured release plan, moves it to Ready for Review, then stops — it does NOT execute the publish command.

This skill should be used when the user asks to "publish a plugin", "release a plugin", "bump plugin version", "update a Claude Code plugin", "update a Codex plugin", "publish skills", or mentions plugin publishing, plugin release, or skill distribution. Handles synchronized host manifests, changelog and README updates, git workflow, cache refresh, and standalone Agent Skills.

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.

Capture writing style profiles, track a pool of users, scan social intelligence, and apply style-matching to draft content. Use when asked to "capture my writing style", "draft a post in my voice", "scan what's trending", "add someone to the pool", or "track @username".

Explore a running app on randomized unscripted paths to surface new bugs, broken flows, and confusing UX, then file deduplicated tickets. Use for "free roam", "explore the app", "monkey test", "exploratory testing", "find issues I don't know about", or "discovery loop". Respects a never-touch list and blast-radius boundary. Not for scripted regression tests.

This skill should be used when the user asks to 'find bugs', 'do a thorough code review', 'run a security audit', 'hunt for bugs', 'check for correctness issues', or 'review this code for edge cases'. Orchestrates a three-phase adversarial review using three isolated agents — Jerry (Hunter), Kayle (Skeptic), Jason (Referee) — to neutralize sycophancy and produce high-fidelity bug reports. User-facing command: /bug-hunt

Produce one self-contained, theme-aware HTML page led by grounded diagrams for work that has not been built. Use for "make a visual proposal", "present these options visually", "diagram the trade-offs", "turn this plan into something reviewable", or a shareable design pitch, RFC, or options comparison. Use visual-review for completed code changes.

Turn a PR, branch, commit, or working-tree diff into one self-contained HTML recap page with before/after wireframes, contract summaries, a file map, and annotated diffs. Use for "recap this PR", "visual recap", "review this diff visually", "show me what this branch changed", "critique this change", or before any large or UI-heavy review. Not for small single-file diffs.

Data visualization and charting. Use for "create a chart", "visualize this data", "build a dashboard", "plot this", "graph these metrics", "make a bar chart", "build a heatmap", or help choosing a chart type, picking a charting library by data scale, or building the pipeline from raw database state to rendered visualization.

This skill should be used when the user asks to "inspect this page in Chrome", "debug what's happening in my browser", "read the browser console logs", "interact with the page open in Chrome", or "attach to my live Chrome session". Interacts with the user's local Chrome browser session via Chrome DevTools Protocol — no extension, no Puppeteer — but only after explicit user approval to inspect, debug, or interact with an open tab.

This skill should be used when the user asks to "create a new project", "scaffold a Next.js app", "initialize a new app", "start a new project", "set up a new Next.js project", or mentions "create-next-project". Provides a guided, opinionated full-stack Next.js project initialization with Biome, Tailwind v4, shadcn/ui, better-auth, and Vercel deployment. Uses agent teams for parallel execution.

This skill should be used when the user asks to "optimize my Next.js app", "why is this page slow", "reduce my bundle size", "improve Core Web Vitals", or "run a Lighthouse audit". Diagnoses and fixes Next.js frontend performance issues using Lighthouse, bundle analysis, and animation best practices — covering LCP, TBT, and CLS.

Decide between static components and AI-generated interfaces, and build the latter. Use for "generative UI", "dynamic UI", "json-render", "render JSON as UI", "AI UI generation", "MCP App UI", "deliver UI in chat", or "interactive chat interface". Covers the json-render framework, renderer and catalog selection, and MCP Apps delivery through ui:// resources.

This skill should be used when adding GitHub star counts, star buttons, star widgets, or GitHub social proof to a website or app. Applies when the user says "add GitHub stars", "show star count", "add a star badge", "GitHub badge", "star widget", "GitHub social proof", "stargazer count", or wants to display how many stars a repo has on a marketing page, header, or landing page. Also applies when integrating the GitHub API for repository metadata display.

Assess and upgrade a Next.js project after resolving the target release from current official documentation. Includes tested automation and breaking-change guidance for the Next.js 15 to 16 migration. Use when the user says 'upgrade Next.js', 'migrate to Next.js 16', 'update my Next.js app', 'run the Next.js codemod', 'my Next.js version is outdated', or when a migration plan or baseline build comparison is needed.

Run local performance audits on a project without network calls. This skill should be used when the user says 'audit performance', 'check bundle size', 'find large images', 'check for heavy dependencies', 'run a perf audit', 'how big is my bundle', 'optimize images', 'find oversized assets', or before any performance optimization work. Also use when an agent needs baseline metrics before making changes. All scripts output structured JSON to stdout.

This skill should be used when the user asks to "audit my shadcn app", "run shadscan", "improve my shadscan score", "raise my shadscan grade", "fix accessibility/states/forms regressions in my shadcn components", or "add a shadscan CI gate". Drives the shadscan static analyzer (`npx @shadscan/cli`) through an audit → fix → re-score loop against shadcn React components, and wires it into CI as a score gate.