Find the capability.
Keep its context.
Every skill stays connected to its publisher, source plugin, runtime support, companions, and agents.
advisor
Active when a Claude Code or Codex main session needs an independent, read-only second opinion at a commitment boundary. Use before substantive work on a hard task, when stuck or changing approach, at a final review gate, or when the user says "consult the advisor", "get a second opinion", "ask codex", "ask Fable", "ask a bigger model", or wants an advisor set up. Supports Claude-native advisor behavior, Codex-as-advisor, and a Codex-main to Claude Fable CLI channel. The advisor returns guidance; the main session retains execution and decision ownership.
agent-auditor
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.
agent-decommissioning
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).
agent-onboarding
Complete end-to-end checklist for adding a new agent to the bOpen team. Use when creating a new agent, onboarding a new team member, or need to remember the full agent deployment pipeline — design, write, avatar, plugin, Paperclip registration, roster, and optional ClawNet bot deployment.
benchmark-skills
Use this skill when creating evals or assertions for a skill, running the skill benchmark harness, measuring skill effectiveness vs baseline, or writing evals.json files alongside skills. Invoke whenever someone asks to test, benchmark, or evaluate a skill's quality.
charting
Full-stack data visualization and charting intelligence. This skill should be used when the user asks to "create a chart", "visualize this data", "build a dashboard", "plot this", "graph these metrics", "show me a chart of", "make a bar chart", "create a line graph", "build a heatmap", or needs help choosing the right chart type, selecting a charting library, or engineering the data pipeline from raw database state to rendered visualization. Covers chart selection, data transformation, library choice by scale, performance optimization, and accessibility.
check-version
Check if the bopen-tools plugin is up to date by comparing local vs GitHub versions. Use this skill at the start of any session where the agent needs current skill/agent definitions, when the user asks 'is everything up to date?', 'check for updates', 'am I on the latest version?', or when you suspect skills may have changed since last install. Also use proactively when a skill seems to behave differently than documented or when instructions reference features you don't recognize. Completes in under 100ms.
chrome-cdp
Interact with local Chrome browser session (only on explicit user approval after being asked to inspect, debug, or interact with a page open in Chrome). Connects via Chrome DevTools Protocol — no extension, no Puppeteer, handles 100+ tabs.
clawnet-cli
Reference for ClawNet CLI internals, architecture, and recent changes. Use this skill when working on clawnet, clawnet-paperclip-plugin, or any code that interacts with the ClawNet registry, vault, ORDFS content fetching, or agent/organization publishing.
cli-demo-gif
Generate CLI demo GIFs using vhs (Charmbracelet). Use when creating terminal recordings for README files or documentation.
code-audit-scripts
Run deterministic code security and quality scans — secret detection, debug artifact cleanup, and TODO/FIXME tracking. Use this skill before any security review, code audit, PR review, or when the user says 'scan for secrets', 'find debug logs', 'check for TODOs', 'audit this code', 'security scan', or 'clean up before shipping'. Also use proactively before deployments or when reviewing unfamiliar codebases. Runs all scans in parallel for speed.
codex-agent-setup
Explicit-only installer for bopen-tools Codex custom agents. Use ONLY when the user explicitly asks to install, update, check, uninstall, or set up Codex agents / custom agents from this plugin — phrases include "install codex agents", "setup codex agents", "update codex agents", "check codex agents", "install all codex agents", "uninstall codex agents", "project codex agents", "user codex agents", or "install bopen agents into .codex". Never auto-invoke. Never silently modify global Codex configuration.
confess
Invoke this skill BEFORE ending any session, marking a task done, or saying 'complete'. Also invoke when the user says 'are you sure?', 'did you miss anything?', 'anything else?', 'what did I miss?', 'confess', or 'audit your work'. This is a proactive self-audit — do not wait to be asked. Skipping this means shipping incomplete work, broken references, untested paths, or unmentioned concerns the agent noticed but buried. Covers: incomplete changes, untested assumptions, pattern violations, hidden concerns, and cleanup debt.
coordinator
Always active when a capable current main session in Claude Code or Codex is planning non-trivial implementation and cheaper or specialized executors are available, including native subagents, Grok workers, Codex workers, or lower-tier Claude agents. Trigger before code is written to decide where each unit executes. Keep planning, specs, design intent, review, verification, and git in the main seat; dispatch bounded code-writing volume with explicit ownership and acceptance criteria.
cost-tracking
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".
create-next-project
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.
critique
This skill should be used when the user asks to "show me the changes", "what did you change", "let me see the diff", "review the code", "open web preview of changes", "show me what you worked on", "compare branches", "explain what changed", or wants to view git diffs in a terminal UI, web browser, or get AI-powered code review explanations.
deploy-agent-team
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 bopen-tools specialist roster.
devops-scripts
Deterministic shell scripts for infrastructure health checks and environment validation. This skill should be used when checking deployment health, verifying service connectivity, validating required environment variables before deployment, running pre-deploy smoke checks, diagnosing connectivity issues with Vercel, Railway, Redis, or PostgreSQL, or when the agent needs structured JSON output about infrastructure state without burning context on inline bash logic.
ezkl
This skill should be used when the user asks about "zero-knowledge ML", "zkML", "EZKL proofs", "prove ML inference", "verify model output", "ZK proof for machine learning", "ONNX to ZK circuit", "on-chain ML verification", "EVM verifier for ML", or needs to generate, verify, or deploy zero-knowledge proofs for machine learning models. Also use when working with @ezkljs/engine, ezkl CLI, or Lilith managed proving.
free-roam-testing
Use this skill to run a discovery loop that explores a running app the way a curious or chaotic human would — randomized, unscripted paths, weird inputs, edge interactions — to surface NEW bugs, broken flows, confusing UX, and slow spots, then file them as deduplicated tickets. Invoke it when the user says "free roam", "explore the app", "monkey test", "find issues I don't know about", "exploratory testing", "use it like a real user", "discovery loop", "surface new bugs", or wants an agent to poke around an app autonomously and report what's broken. This is the producer half of the loop architecture — it feeds tickets to the execution loop. NOT for scripted regression tests (that's the tester agent's verification gate). Always respects a never-touch list and a blast-radius boundary before mutating anything.
front-desk
This skill should be used in Claude Code or Codex when the user asks "who handles X?", "what agents are available?", "how do I contact Y?", "team roster", "what services do we use?", "who should I talk to about Z?", "what skills are available?", "where do I find skill X?", or needs help routing to the right agent or service provider. Also use when checking agent-adapter availability, finding/installing skills, sending emails on behalf of the org, or drafting communications. Route SOC 2, audit readiness, policy drafting, and evidence-gathering questions to Anthony in product-skills, with Paul in bopen-tools for technical control validation.
frontend-performance
This skill should be used when the user wants to optimize Next.js frontend performance using Lighthouse, bundle analysis, and animation best practices. Use when diagnosing slow pages, optimizing bundle size, or improving Core Web Vitals (LCP, TBT, CLS).
generative-ui
This skill should be used when the user asks about "generative UI", "dynamic UI", "AI-generated interfaces", "json-render", "render JSON as UI", "generate a dashboard", "create dynamic components", "AI UI generation", "MCP App UI", "deliver UI in chat", "interactive chat interface", or needs to decide whether to use static components vs AI-generated UI. Covers the json-render framework, renderer selection, catalog design, MCP Apps delivery (ui:// resources for in-chat interactive UIs), and integration with gemskills for visual asset generation.
geo-optimizer
This skill should be used when the user asks to "audit for AI visibility", "optimize for ChatGPT", "check GEO readiness", "analyze hedge density", "generate agentfacts", "check if my site works with AI search", "test LLM crawlability", "check discovery gap", or mentions Generative Engine Optimization, AI crawlers, Perplexity discoverability, or NANDA protocol.
github-stars
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.
hammertime
This skill should be used when the user mentions a behavioral rule they want enforced, says "always do X", "never do Y", "stop doing Z", "from now on", asks about HammerTime rules, wants to create a stop hook rule, mentions behavioral guardrails, or wants to understand how the HammerTime stop hook system works. Teaches how to write rules for the HammerTime stop hook system.
hook-manager
Discover and install automation hooks for Claude Code and Opencode. This skill should be used when users ask to "list hooks", "install a hook", "show available hooks", "enable hook", "what hooks are available", or need help managing agent automation hooks.
html-to-pdf
This skill should be used when the user asks to "design a business card", "make a printable PDF", "render HTML to PDF", "generate a postcard", "build print collateral", "set up an HTML print pipeline", or needs help with bleed, safe areas, font embedding, or QR generation for print. Provides a Playwright-based pipeline with multiple bundled templates and theme variants for business cards (minimal, watercolor light, watercolor dark) and instructions for adding new templates.
humanize
Invoke this skill whenever producing text that a human will read — emails, messages, documentation, reports, blog posts, announcements, commit messages, or any prose draft. Trigger signals include: task is a writing or editing task, output will be sent or published, user says "humanize", "make this sound less AI", "de-AI this", "this sounds like ChatGPT", "make it sound more natural", or "edit this". Do not wait for the user to ask — apply this automatically before delivering any human-facing draft. Skipping this means delivering text with predictable AI patterns that erodes trust and sounds generated.
hunter-skeptic-referee
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
linear-planning
This skill should be used when the user wants to plan a project, feature, or bug fix using Linear as the agent control plane. Use when the user says "plan this in Linear", "create Linear tickets", "break this down into tasks", "push to Linear", "set up our board for this feature", or wants to turn a description or spec into well-structured, agent-ready Linear issues. Requires the official Linear MCP server to be configured.
loop-engineering
Use this skill whenever designing, configuring, or hardening an autonomous agent "loop" — a goal an agent iterates toward on its own with a real verification gate, persistent state, and a stop condition. Invoke it when the user mentions "build a loop", "agentic loop", "self-iterating agent", "run this on a schedule/cron", "/loop or /goal", "Ralph loop", "maker-checker", "fleet of agents", "autonomous workflow", or wants an agent to keep working a goal unattended until it's verifiably done. Also use when scoping whether a loop is even worth building, when picking a verification gate, when deciding what a loop is allowed to touch (blast radius), or when a loop is burning tokens without producing accepted work.
mcp-apps
This skill provides guidance for building MCP Apps, the official MCP extension (io.modelcontextprotocol/ui) for rendering interactive HTML UIs inside MCP hosts. This skill should be used when the user asks to "create an MCP App", "add UI to MCP tool", "build interactive MCP", "MCP App server", "ui:// resource", "sandboxed iframe MCP", "interactive chat UI", "embed UI in chat", "MCP tool with interface", or needs to build interactive HTML applications that render inside Claude Desktop, ChatGPT, or VS Code Copilot.
nextjs-upgrade
Upgrade a Next.js project to the latest version (v16) with Turbopack, async Dynamic APIs, Biome, and React 19.2. This skill should be used 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 is needed before making Next.js changes. Also invoke when the agent needs baseline build metrics for before/after comparison, or when checking whether a project needs async API migration, middleware-to-proxy migration, or Biome adoption.
notebooklm
This skill should be used when the user wants to query Google NotebookLM notebooks directly from Claude Code for source-grounded, citation-backed answers from Gemini. Provides browser automation, library management, and persistent auth. Drastically reduced hallucinations through document-only responses.
npm-publish
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).
orchestrator
Use this skill when a capable current Claude Code or Codex main session should coordinate native specialist agents, external implementation workers such as Grok, and an independent advisor such as Fable. Trigger for "orchestrate this", "use Grok workers", "use Fable as advisor", "Codex main with workers", "delegate implementation but keep control here", cross-model workflows, or complex work that needs a main-seat plan, parallel specialists, worker dispatch, second opinions, review, verification, and git ownership. Do not hardcode or replace the user's current main model.
paperclip-plugin-dev
Build, publish, and install Paperclip plugins correctly. This skill should be used when scaffolding a new Paperclip plugin, writing a plugin manifest, implementing plugin worker logic, adding UI slots, publishing to npm, or installing a plugin into a Paperclip instance. Contains critical lessons from real publishing failures. Also invoke when working on plugin capabilities, jobs, webhooks, agent tools, or the plugin SDK.
perf-audit
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.
persona
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".
plaid-integration
This skill should be used when the user wants to integrate Plaid API for bank account connections and transaction syncing. Use when implementing financial data access, bank linking, or transaction imports in TypeScript/Bun applications.
process-cleanup
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.
publish-request
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.
reinforce-skills
Invoke this skill when: modifying any CLAUDE.md file, adding a new skill or agent to a plugin, user says 'update the skill map', 'add this to the map', 'register this agent', 'skills keep getting forgotten', 'I keep forgetting which skill to use', 'agents keep getting forgotten', 'add skill map', 'update agent map', 'sync skills to CLAUDE.md', or when setting up a new project. This skill injects compressed SKILL-MAP and AGENT-MAP directive blocks into CLAUDE.md so skill names and agent IDs persist across the session without fading from context. Skipping this means agents will forget skill names mid-session, fail to invoke the right skill, and guess at agent IDs — causing silent capability loss that is hard to diagnose.
remind
Search and recall previous Claude Code conversation sessions. Use this skill whenever the user asks to remember, recall, find, or look up something from a past conversation or session. Triggers on phrases like: 'remember when we...', 'what did we do about...', 'find that conversation where...', 'when did I last work on...', 'what was that command/approach/solution we used for...', 'look up my past sessions about...'. Also use this skill when the user references prior work context that isn't in the current session, asks to continue work from a previous session, or wants to find a specific discussion, decision, or code snippet from their conversation history. Even vague references to past work ('that thing we did', 'the approach from last week') should trigger this skill.
runtime-context
Detects agent execution environment (Claude Code, Vercel Sandbox, or local dev) and adapts behavior accordingly. This skill should be used when an agent or bot needs to understand what runtime it is in, what tools are available, or how to adapt its behavior across different execution contexts. Use this skill when building agents that may run in Claude Code as subagents AND as hosted bots in Vercel Sandboxes, or when a SOUL.md/SKILL.md needs to work across runtimes.
saas-launch-audit
This skill should be used when the user asks to "audit my SaaS", "check if I'm ready to launch", "review my launch checklist", "verify my pricing", "audit my payment setup", "check my AI visibility", "prepare for Product Hunt", "validate my SaaS for launch", or mentions launching a SaaS product. Provides a comprehensive, repeatable checklist with PASS/FAIL verification and actionable next steps.
shaders
This skill should be used when writing custom shaders for Three.js, creating visual effects with GLSL or TSL (Three Shader Language) for WebGL and WebGPU, debugging shader issues, building post-processing pipelines, implementing noise functions, procedural textures, or custom materials. Covers shader workflow, TSL node system, GLSL patterns, debugging, performance optimization, and post-processing with pmndrs/postprocessing.
skill-publish
This skill should be used when the user asks to "publish a plugin", "release a plugin", "bump plugin version", "update a Claude Code plugin", "publish skills", or mentions plugin publishing, plugin release, or skill distribution. Handles version bumping, changelog updates, git workflow, and publishing for both Claude Code plugins and standalone Agent Skills.
statusline-setup
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.
threejs-r3f
This skill should be used when building Three.js or React Three Fiber (R3F) projects, creating 3D scenes, animating meshes with useFrame, loading GLTF/GLB models, setting up physics with @react-three/rapier, using WebGPU with R3F, optimizing 3D performance, scaffolding Vite+R3F projects, or exporting R3F components. Covers scene setup, Drei helpers, asset pipeline, responsive canvas, and performance budgets.
ui-audio-theme
Generate cohesive UI audio themes with subtle, minimal sound effects for applications. This skill should be used when users want to create a set of coordinated interface sounds for wallet apps, dashboards, or web applications - generating sounds mapped to UI interaction constants like button clicks, notifications, and navigation transitions using ElevenLabs API.
visual-recap
Use this skill to turn a PR, branch, commit, or working-tree diff into a visual recap — a single self-contained, theme-aware HTML page with before/after UI wireframes, schema/API contract summaries, a file footprint map, and annotated key-change diffs — so a reviewer sees the SHAPE of a change before reading raw lines. Invoke it when the user says "recap this PR", "visual recap", "show me what this branch changed", "summarize this diff visually", "make this PR reviewable", "explain what changed", or before reviewing any large, multi-file, UI-heavy, or schema/API-touching change. Code-review agents (code-auditor, architecture-reviewer) should produce one as the opening artifact of any big review. NOT for small single-file diffs — those review faster as a raw diff via the critique skill.
voice-clone
Clone real or fictional voices using ElevenLabs Instant Voice Cloning (IVC). This skill chains together the full pipeline — finding reference audio, preparing samples, uploading to ElevenLabs IVC, testing the clone with text-to-speech, and tuning voice settings. Use this skill whenever the user wants to clone a voice, create a custom voice from audio samples, replicate a famous voice style, or build a voice for a character. Covers celebrity impressions, fictional characters, branded voices, and personal voice clones.
wait-for-ci
Wait for CI/CD pipelines to complete after pushing code, then act on results. This skill should be used after git push, after creating a PR, when the user says 'wait for CI', 'check if the build passes', 'monitor the pipeline', 'wait for checks', 'is CI green?', or whenever the agent needs to verify that pushed code passes CI before proceeding. Also use when an agent workflow involves push-then-verify cycles, deployment monitoring, or needs to block on CI results before taking the next step. Supports GitHub Actions, GitLab CI, and Vercel deployments.
wave-coordinator
This skill should be used in Claude Code or Codex when dispatching more agents than the host can run concurrently, when context budget management is needed, or when generating multiple variations of the same output. Coordinates native Claude agents, Codex custom or built-in subagents, and mixed worker lanes with wave sizing, context budget tracking, and directive diversity. Use when the user says "fan out", "generate variations", "batch agents", "wave dispatch", or asks for large-scale subagent work.
x-research
AI-powered X/Twitter research via xAI Grok. Returns AI SUMMARIES with analysis, not raw tweets. Use for "what's trending", "social sentiment", "summarize X discussion about", "analyze X conversation about", "research topic on X". For RAW tweet data, use x-user-timeline, x-tweet-search, x-tweet-fetch instead. Requires XAI_API_KEY.
x-tweet-fetch
Fetch a specific tweet by URL or ID. Use when user shares an X/Twitter URL (https://x.com/... or https://twitter.com/...), asks "get this tweet", "fetch tweet", "what does this tweet say", "read this X post". Requires X_BEARER_TOKEN.
x-tweet-search
Search recent X/Twitter posts by query. Returns RAW TWEETS (last 7 days). Use when user asks "search X for", "find tweets about", "what are people saying about", "Twitter search", "raw tweets about". For AI summaries/sentiment, use x-research instead. Requires X_BEARER_TOKEN.
x-user-lookup
Get X/Twitter user profile by username. Use when user asks "who is @username", "get X profile", "lookup Twitter user", "find X account", "user details", "follower count for". Requires X_BEARER_TOKEN.
x-user-timeline
Get recent tweets from an X/Twitter user. Use when user asks "what has @username posted", "recent tweets from", "user's X posts", "show timeline for", "what is @user saying". Requires X_BEARER_TOKEN.