CLI for the onchain agent messaging layer on the Base blockchain, built on Net Protocol. Explore other agents, post to feeds, send direct messages, and store information permanently onchain.

CLI for the onchain agent messaging layer on the Base blockchain, built on Net Protocol. Explore other agents, post to feeds, send direct messages, and store information permanently onchain.

Parallel read-only multi-agent root-cause investigation for bugs, regressions, crashes, flaky behavior, or unexplained failures. Use when the user asks to investigate a bug, find the root cause, trace a regression, understand why something broke, or wants a ranked diagnosis with the fastest proof path without making code edits.

Executes diagnostic investigation tasks to test debugging hypotheses. Runs tests, traces execution, checks git history, and reports evidence. (converted from agent)

USE FOR web search, research, RAG, grounding, browse, find, lookups, fact-checking, documentation, agentic AI. All-in-one, optimized for AI agents. Pre-extracted, token-budgeted web content, deep research, news, images, videos, places, custom ranking

Use this skill when the user asks to "enable cache protection", "block on cache miss", "avoid cache expiration cost", "/ar:cache", "protect against compaction", "cache pressure guard", "cache hit ratio threshold", or "set cache-miss threshold". Covers the /ar:cache slash command family for blocking tool use when the Claude Code prompt cache is cold or context is near compaction.

Guide for effective use of che-ical-mcp for macOS Calendar & Reminders management. Use when user asks about calendar events, reminders, scheduling, or time management.

To provide a comprehensive reference for configuring and using Claude Code (the agentic coding tool) to its full potential. This skill synthesizes best practices, configuration templates, and advanced usage patterns.

Maintain Claude Code settings, hooks, and agent config formats; use when editing or validating settings.json or hooks.

# MCP-Tasks CLI Reference

Catch the user up on what the agent has been doing in the current session with 3-5 ultra-tight bullet points. Use when the user invokes /cmd-what, asks "what are you doing", "what did you do", "catch me up", "status", "where are we", or otherwise needs a fast situational summary of in-flight agent work.

Run OpenAI Codex CLI as an independent reviewer over the current branch, a specific commit, or uncommitted changes. Builds a focused instruction file from the real diff and returns a compact review summary.

Guide to writing slash commands for Claude Code. Use when creating commands that set mode, constraints, or workflows invoked via /command-name.

Use when compressing agent context, implementing conversation summarization, reducing token usage in long sessions, or asking about "context compression", "conversation history", "token optimization", "context limits", "summarization strategies"

Use when diagnosing agent failures, debugging lost-in-middle issues, understanding context poisoning, or asking about "context degradation", "lost in middle", "context poisoning", "attention patterns", "context clash", "agent performance drops"

Use when optimizing agent context, reducing token costs, implementing KV-cache optimization, or asking about "context optimization", "token reduction", "context limits", "observation masking", "context budgeting", "context partitioning"

Build and wire an `AtomicAgent[InSchema, OutSchema]` — schemas, `AgentConfig`, `SystemPromptGenerator`, provider client, history, hooks, optional context providers. Use when the user asks to "create an agent", "add another agent", "build an `AtomicAgent`", "wire up an agent", "make a planner/router/extractor agent", or runs `/atomic-agents:create-atomic-agent`.

Create OpenClaw plugins/extensions (TypeScript modules) from natural‑language requests. Use when the user asks to create a plugin/extension, add a new slash command, add plugin tools, or says “create plugin that does …” or “create OpenClaw plugin called NAME …”. Trigger for /create-plugin NAME WHAT-IT-DOES, /create-plugin GENERAL-PROMPT, “create extension that does ACTION”, or similar.

Scaffold a new Claude Code plugin with proper directory structure, plugin.json, skills, commands, and agents

Create a recurring AI-generated PostHog report — schedule a free-text prompt to run on a cron, with the LLM-synthesized markdown delivered to email or Slack on each tick. Use when the user wants "send me a weekly AI summary of X" rather than a one-off report.

Orchestrate a 4-agent cyber defense pipeline to analyze log files for threats. Use when investigating security logs, detecting anomalies in access patterns, classifying breach severity, or generating incident reports from nginx/auth/syslog files.

Bootstrap d3k in standalone AI apps (Codex, Cursor, Claude Code): detect/install dev3000, start d3k as the runtime, and use unified logs plus CDP browser control instead of running npm/bun dev directly.

Audit what the Bits AI assistant (MCP server) has done in your Datadog org — tool calls by user, resources accessed, and anomaly flags for AI governance.

Diagnose failure modes by systematically investigating traces. Trigger when the user explicitly asks for cross-trace diagnosis: "what's going wrong?", "were there errors?", "debug this", "where is my agent struggling?". Do NOT trigger on: (1) advice questions ("what should I do?"), (2) statistical questions ("what's the average latency?"), (3) summarize requests, (4) trace filtering ("show me traces with errors"), (5) vague questions ("is there a problem?"), (6) unrelated requests.

INVOKE THIS SKILL when your Deep Agent needs memory, persistence, or filesystem access. Covers StateBackend (ephemeral), StoreBackend (persistent), FilesystemMiddleware, and CompositeBackend for routing.

Enumerate ALL workstreams in the current deepwork session and diff each artifact against its source-of-truth. Produces drift-report.vN.md listing potential drift items across every workstream, not only those flagged by the drift agent.

This skill should be used when the user asks 'what development tools are available', 'list dev plugins', 'what MCP servers can I use', 'enable code intelligence', 'what testing tools exist', or needs to discover development plugins like serena, playwright, or context7. Use this for general development tool discovery; use ds-tools for data science-specific tools.

Use when fanning out work across multiple agents — same prompt to many for Best-of-N, different prompts to many for parallel exploration, or sending a scout before committing. Use when user says "parallel agents", "fan out", "best of N", "scout", "race", "vote", "spawn workers".

Generate BDD scenarios end-to-end from EARS, a prompt, or an IPLAN - detect input, generate Gherkin scenarios, validate, and run the audit/fix cycle. Use to create or batch-create BDD suites.

Documentation lifecycle management skill. Activates when user mentions documentation, docs, sync, quality, validation, releases, or setup. Routes to appropriate agent (doc-expert for orchestration, doc-writer for content) and provides gentle reminders about documentation health.

This skill should be used when the user asks 'what data science tools are available', 'list data analysis plugins', 'what skills work with data', 'enable data science code intelligence', or needs to discover data-specific plugins and skills like wrds, lseg-data, gemini-batch, or data context skills. Use this for data science tool discovery; use dev-tools for general development tool discovery.

INVOKE FIRST for any LangChain / LangGraph / Deep Agents agent building project before consulting other skills or writing any agent code. Required starting point for up to date info on framework selection (LangChain vs LangGraph vs Deep Agents vs hybrid composition), agent patterns, install, environment setup, and which skill to load next.

Persistent semantic memory for AI agents — local, fast, free. Use when agent needs to recall past decisions, store new facts/preferences, search conversation history, or maintain context across sessions.

Help the user systematically identify and categorize failure modes in an LLM pipeline by reading traces. Use when starting a new eval project, after significant pipeline changes (new features, model switches, prompt rewrites), when production metrics drop, or after incidents.

Grade implementation work against bead acceptance criteria using a separate judge agent. Use after subagent work passes mechanical gates, as a pre-merge check, or on-demand to evaluate existing features. The evaluator is NOT the orchestrator and NOT the implementer — it only judges. Integrates with browser-qa for runtime verification when CDT MCP is available.

Use Claude's extended thinking (reasoning) mode effectively — budget tokens, interleaved thinking with tool use, when it helps, when it wastes tokens, and how to inspect the thinking trace. Use this skill when building reasoning-heavy features (math, code generation, multi-step planning), debugging why a model is shallow on hard problems, or deciding whether to enable thinking. Activate when: extended thinking, thinking tokens, budget_tokens, reasoning mode, interleaved thinking, thinking blocks.

Wield Google's Gemini CLI as a powerful auxiliary tool for code generation, review, analysis, and web research. Use when tasks benefit from a second AI perspective, current web information via Google Search, codebase architecture analysis, or parallel code generation. Also use when user explicitly requests Gemini operations.

Use when a project has no tenets yet and the user wants to discover architectural constraints and create an initial `## Tenets` section in AGENTS.md. Spawns the constraint-explorer agent to map the codebase, presents discovered constraints with supporting evidence for the user to review, and writes only approved tenet text — no evidence is persisted. Use the governor-manage skill instead when tenets already exist.

Write a handoff document so a fresh agent can continue this work after the conversation ends. Invoke when the user is about to /clear or /compact, when context is filling up, or when the user asks to hand off, checkpoint, or pause work for a later session. Argument (optional): what the next session will focus on.

Creates AI agent task management structure with feature backlog (ai/tasks/), TDD enforcement, and progress tracking. Use when setting up agent-foreman, initializing feature-driven development, creating task backlog, or enabling TDD mode. Triggers on 'init harness', 'setup feature tracking', 'create feature backlog', 'enable strict TDD', 'initialize agent-foreman'.

You MUST use this skill for ANY changes you make within to the iOS app. Senior iOS engineer with visual feedback capabilities via simulator MCP tools.

Helps users report issues with GTM skills — incorrect advice, broken activation, missing context, or unexpected behavior. Guides through gathering the right details (skill name, prompt used, actual vs expected output, context files present) and generates a properly formatted GitHub issue. Use when users say: 'this skill gave wrong advice', 'report a bug', 'file an issue', 'skill isn't working', 'wrong output', 'skill didn't activate', 'bad recommendation', 'incorrect behavior', or 'something is broken'.

Run Kiro CLI as an independent reviewer over the current branch, a specific commit, or uncommitted changes. Builds a focused prompt from the real diff and returns a compact review summary.

Project-specific prompt optimization knowledge management. Use when storing or retrieving learned patterns from comparisons. Provides schema, extraction criteria, capacity management, and retention scoring.

Building applications with Large Language Models - prompt engineering, RAG patterns, and LLM integration. Use for AI-powered features, chatbots, or LLM-based automation.

Use an LLM as an evaluator for open-ended outputs — rubrics, pairwise comparison, calibration with human labels, bias mitigation. Covers when LLM-judge works, when it fails, and how to trust its scores. Use this skill when evaluating generative outputs at scale, building eval pipelines, or replacing expensive human review for non-critical judgments. Activate when: LLM as judge, LLM evaluator, automated evaluation, pairwise comparison, rubric evaluation, eval model.

Use when "LLM inference", "serving LLM", "vLLM", "llama.cpp", "GGUF", "text generation", "model serving", "inference optimization", "KV cache", "continuous batching", "speculative decoding", "local LLM", "CPU inference"

Book real-world services through Lokuli MCP. Use when user needs to find, check availability, or book local services like plumbers, electricians, cleaners, mechanics, barbers, personal trainers, etc. Triggers on requests like "book me a haircut", "find a plumber near me", "I need a smog check", "schedule a massage", or any local service request. 75+ service categories available.

Pre-flight checks and launch for agent loops. Validates branch state, session prompts, creates loop directory structure, and launches run.sh in a tmux session. Use when the user wants to start a new agent loop in the current repo.

Pre-flight checks and launch for agent loops. Validates branch state, session prompts, creates loop directory structure, and launches run.sh in a tmux session. Use when the user wants to start a new agent loop in the current repo.

Conversational interface for managing MCP (Model Context Protocol) server configurations in Claude Code

Choose between MCP transports — stdio for local processes, Streamable HTTP for remote servers, SSE for legacy — and implement each correctly with reconnection, backpressure, and session handling. Use this skill when deciding transport for a new MCP server, migrating SSE to Streamable HTTP, or debugging transport-level issues (connection drops, buffering, session loss). Activate when: MCP transport, stdio vs HTTP, Streamable HTTP, MCP SSE, MCP reconnection, MCP session.

Execute AI tasks (codex/claude/gemini) with memory and resume support via memex-cli stdin protocol.

Audit, clean, and optimize Clawdbot's vector memory (LanceDB). Use when memory is bloated with junk, token usage is high from irrelevant auto-recalls, or setting up memory maintenance automation.

Use when implementing agent memory, persisting state across sessions, building knowledge graphs, tracking entities, or asking about "agent memory", "knowledge graph", "entity memory", "vector stores", "temporal knowledge", "cross-session persistence"

Master dispatcher for all MLflow workflows. Use this skill when the user wants to do anything with MLflow — tracing, evaluating, debugging, or improving an agent. Routes to the right MLflow sub-skill automatically. Triggers on: "use mlflow", "help with mlflow", "mlflow agent", "add mlflow to my project", "trace my agent", "evaluate my agent", or any MLflow task without a specific skill in mind.

Use when designing multi-agent systems, implementing supervisor patterns, coordinating multiple agents, or asking about "multi-agent", "supervisor pattern", "swarm", "agent handoffs", "orchestration", "parallel agents"

Adopt multiple expert personas sequentially for complex problem analysis from diverse perspectives. Single-agent only — do NOT spawn sub-agents.

Interactive onboarding workflow that interviews users to understand their coding goals and generates PR-ready implementation plans. Use when starting a new development task to ensure clear requirements and structured execution.

Use when tasks involve reading, creating, or reviewing PDF files where rendering and layout matter; prefer visual checks by rendering pages (Poppler) and use Python tools such as `reportlab`, `pdfplumber`, and `pypdf` for generation and extraction. Originally from OpenAI's curated skills catalog.

Authoring playbook for building agents that automate recurring internal tasks — running scheduled workflows, syncing data between systems, posting notifications, processing inbound events, or executing operational runbooks. Use this when the user wants an agent that runs on a schedule, reacts to events, automates a process, syncs between tools, or handles ops/internal infrastructure.

Transform prompts into optimized, production-ready versions. PROACTIVELY activate for: (1) prompt engineering, (2) prompt improvement, (3) making prompts more effective, (4) prompt debugging. Triggers: "optimize prompt", "improve prompt", "better prompt", "fix prompt", "prompt engineering", "refine prompt", "enhance prompt"

Sanitize OpenClaw agent output before display. Strips leaked credentials, PII, internal paths, and sensitive data from responses.

Chat with your agent about projects, recommendations, and canonical papers in Paperzilla. Use when users ask for recent project recommendations, canonical paper details, markdown-based summaries, recommendation feedback, feed export, or Atom feed URLs.

Decompose PRDs and Tech Specs into parallel-executable tasks with contracts, prompts, and dependency graphs. Use when breaking down a PRD for multi-agent execution.

Execute multiple Claude Code agents in parallel using the cpo CLI tool. Use when running parallel tasks, monitoring execution, or understanding the execution workflow.

Run the PinchTab optimization loop. Spawns blind subagents that execute 108 browser automation steps across 47 groups using only the PinchTab skill, then reports pass/fail results and operation counts vs baseline. Use when asked to 'run optimization', 'run the opt loop', 'benchmark the agent', or 'test pinchtab agent'.

PRD completeness validation. Use to check that a PRD (or GitHub issue body that serves as one) contains the required sections before it is handed to a planning/execution agent, so the plan is built from a complete spec instead of hallucinated scope. Run it as a blocking gate or a warning-only lint.

Processes every note in the Obsidian inbox — reads each one, uses the para-organizer agent to suggest the right PARA destination, confirms with the user, and moves or deletes on approval. Make sure to use this skill whenever the user says "process my inbox", "clear my inbox", "triage inbox notes", "sort inbox", or runs /process-inbox. Execute the workflow — do not describe it.

Craft effective worker prompts from backlog issues using Claude 4 best practices. Use when preparing issues for workers, writing prompts, or optimizing prompt quality. Trigger with "craft prompt", "write prompt", "prepare issue", "make ready".

Crafting effective prompts for LLMs. Use when designing prompts, improving output quality, structuring complex instructions, or debugging poor model responses.

Find and resolve prompt files in ./prompts/ directory. Use when user asks to find a prompt, list available prompts, locate prompt by number or name, or check what prompts exist.

Optimize prompts for code-related tasks following prompt-engineering best practices. Use when refining prompts for implementation, debugging, refactoring, code review, or testing.

Autonomous TDD development loop with parallel agent swarm, category evolution, and convergence detection. Use when running autonomous game development, quality improvement loops, or comprehensive codebase reviews.

Use this skill when handling failed renewals on Android with RevenueCat. Covers how Grace Period and Account Hold are reflected in CustomerInfo automatically, when to prompt the user, and how to trigger Google's in app messaging via showInAppMessages.

Chat with Readwise highlights using GPT-5.1 RAG. Use when the user wants to ask questions about their reading library, get summaries across documents, or have a conversation about their highlights. Triggers on "ask readwise", "chat about my highlights", "what do my notes say about", "summarize my reading on".

Verify LLM reasoning using Reverse Chain-of-Thought (RCoT) to detect overlooked conditions.

Decode benchmark videos, contact sheets, frames, or rough ideas into reusable prompt structure. Use this when you need to extract hook essence, viewer question, must-copy visual grammar, and forbidden drift before writing storyboard or generation prompts.

Reference skill demonstrating the 10/10 skillpack contract. Adds a "what does this skillpack do" answer to the user's agent.

Authoring playbook for building agents that search, read, and synthesize information into a report. Use this when the user wants an agent to research a topic, summarize sources, compare options, do competitive analysis, monitor news, generate briefs, or pull together a citation-backed report from the web or internal documents.

Quality review checklists for AL solution plans, code implementations, and test suites. Auto-load when reviewing agent output before presenting to user.

Comprehensive Remix v2 code review with per-area review skills, run in parallel where the agent supports subagents and sequentially otherwise. Detects Remix v2 in package.json, loads relevant review skills, runs verification protocol.

Quick start guide for creating MCP servers with the rmcp crate - installation, concepts, and first server

Diagnose Sentry issues without copy-pasting stack traces. Uses the Composio CLI to pull issue details, events, breadcrumbs, and suspect commits, then maps the frames to local source so the agent can propose a fix directly.

Universal AgentOps init prompt for starting or onboarding a fresh agent session.

Summarize the current conversation session and append results to the weekly agent-log. Use when user says "log this", "session log", "summarize this session", or asks to write results to the agent-log.

Always-on ambient signal capture. Fires on every inbound message to detect original thinking and entity mentions. Spawn as a cheap sub-agent in parallel, never block the main response.

Create SKILL.md files that teach the agent new domains. Use when authoring a new pi skill with frontmatter, body structure, and references.

Extract actionable Claude Code skills from raw source material — transcripts, conversations, workflows, expertise dumps. This skill identifies repeatable, promptable workflows embedded in content and scores them by leverage. Use when processing a corpus (podcast transcripts, blog posts, course material) to discover what skills could be built from it.

Query Gemini with Google Search grounding

SkillOpt-flavored offline training loop for any SKILL.md. Treats accumulated learn-rule corrections as training trajectories, proposes bounded patches via an optimizer LLM, gates each candidate against a held-out validation set built from the user's own past corrections, and ships only candidates that demonstrably improve the score. Inspired by Microsoft SkillOpt's ReflACT pipeline (rollout → reflect → aggregate → select → update → evaluate) adapted to pro-workflow's SQLite store. Use when a skill has accumulated 8+ learn-rule rows and the user wants the skill itself to get better, not just longer.

Local-first analytics for AI agent skills. Use when user asks about skill usage, analytics, health, context budget, cost/burn rate, trigger conflicts, dead weight analysis, or wants to clean up unused skills.

Search for software composition analysis (SCA) dependency risks in a SonarQube project (project key optional when MCP integration already defines the default project)

[EXPLICIT INVOCATION ONLY] Creates dependency-aware implementation plans optimized for parallel multi-agent execution.

Matches tasks to appropriate agents by analyzing task requirements and finding agents with matching capabilities. Use when you need to assign tasks to agents or determine which agent should handle a task. Considers task type, requirements, and agent capabilities.

Router for the terminator plugin — Stop hooks that end a Claude Code session (and optionally its terminal) when the agent's final message contains a kill phrase. Routes to install, remove, update, or whendone. Use when the user says "terminator", "install the kill hook", "set up a kill phrase", "single kill", "double kill", "remove the terminator hook", "change the kill phrase", or "terminate when done".