Run ML model inference on FiftyOne datasets. Use when running models for detection, classification, segmentation, or embeddings. Discovers available models dynamically from the Zoo, plugin operators, or custom sources — never assumes a fixed model list.

Search, install, list, remove, update, or scaffold AI agents with the `agentshq` CLI across many coding CLIs and IDEs. Use when the user wants to discover agents, install them into specific clients, or manage an existing agent catalog.

Fixes Flutter layout errors (overflows, unbounded constraints) using Dart and Flutter MCP tools. Use when addressing "RenderFlex overflowed", "Vertical viewport was given unbounded height", or similar layout issues.

Pick the right y-axis unit when creating or updating a TrendsQuery insight via `posthog:insight-create` or `posthog:insight-update`. Use when the agent is about to add a `formula` purely to convert units (e.g. dividing seconds by 60 to display minutes), when a `math_property` is a duration, currency, ratio, or large count, or whenever the user mentions "format the y-axis", "duration", "seconds", "minutes", "hours", "milliseconds", "ms", "percentage", "currency", "decimals", "axis label", or "axis unit" in the context of a graph insight.

Use this agent when you need to understand the historical context and evolution of code changes, trace the origins of specific code patterns, identify key contributors and their expertise areas, or analyze patterns in commit history. This agent excels at archaeological analysis of git repositories to provide insights about code evolution and development patterns. <example>Context: The user wants to understand the history and evolution of recently modified files.\nuser: "I've just refactored the authentication module. Can you analyze the historical context?"\nassistant: "I'll use the git-history-analyzer agent to examine the evolution of the authentication module files."\n<commentary>Since the user wants historical context about code changes, use the git-history-analyzer agent to trace file evolution, identify contributors, and extract patterns from the git history.</commentary></example> <example>Context: The user needs to understand why certain code patterns exist.\nuser: "Why does this payment processing...

Expert Gmail management assistant via Rube MCP. Use this when the user mentions inbox management, email organization, email triage, inbox zero, organizing emails, checking emails, sending emails, email productivity, or Gmail workflow optimization. Provides intelligent workflows and best practices for efficient email handling.

Compact the current conversation into a handoff document for another agent to pick up. Use when the user says "hand this off", "wrap up", "save context", "continue later", "pass this to another session", "I need to stop here", "pick this up later", or when a session is ending with unfinished work. Also use when context is getting long and the user wants to start fresh without losing progress.

Structured work handoff across context boundaries: triage decisions, constraints, and remaining work into a prompt-quality transfer document. Invoke whenever task involves handing off work — session restart, teammate delegation, async resumption, or any context boundary crossing.

Author in-repo SKILL.md: frontmatter, validator, structure.

Learn how to enhance your CMS like PocketBase with AI-powered content recommendations using text embeddings, SQLite, and k-nearest neighbor search for efficient and scalable related content suggestions.

Capture feedback about wf, util, or vault plugins and file it as a GitHub issue for improvement. Use after any session where the tooling fell short — wrong agent behavior, missing skill coverage, misleading prompts, workflow friction, or convention gaps. Only for generic plugin improvements that would help ANY project, not repo-specific config. Files to JSai23/claude-tooling with the plugin-feedback label.

Install Chrome DevTools MCP server for debugging and network inspection in Claude Code

Install and configure LSP (Language Server Protocol) for Claude Code to enable go-to-definition, find-references, and real-time diagnostics

Issue creation workflow for Agent Config following the issue-first enforcement system. Trigger: When creating a GitHub issue, reporting a bug, or requesting a feature.

Use when interacting with Jira issues - searching, creating, updating, moving, transitioning, commenting, logging work, downloading attachments, managing sprints, boards, issue links, web links, fields, or users. Auto-triggers on Jira URLs and issue keys (PROJ-123). Also use when MCP Atlassian tools fail or are unavailable for Jira Server/DC.

LinkedIn agent that helps you enrich LinkedIn profiles. You prodive a LinkedIn URL and it will return its data from LinkedIn, in a structured JSON format. Use this skill when the user needs LinkedIn via the bundled Teneo CLI and you need the live commands, arguments, or pricing before execution.

Guide and workflow for LLM Prompt Engineering - System Instruction Best Practices. Use when you need LLM Prompt Engineering - System Instruction Best Practices.

Write high-quality session prompts (SESSION_WORKER.md and SESSION_REVIEWER.md) for an agent loop. Reads repo context and follows the loop-author guide to produce targeted, effective prompts. Use when setting up a new loop or rewriting session prompts for an existing one.

Write high-quality session prompts (SESSION_WORKER.md and SESSION_REVIEWER.md) for an agent loop. Reads repo context and follows the loop-author guide to produce targeted, effective prompts. Use when setting up a new loop or rewriting session prompts for an existing one.

Install, update, list, and remove Claude Code skills. Supports GitHub repositories (user/repo), GitHub subdirectory URLs (github.com/user/repo/tree/branch/path), and .skill zip files. Use when user wants to install, add, download, update, sync, list, remove, uninstall, or delete skills.

Workflow for publishing skills and agents to the dotnet-skills Claude Code marketplace. Covers adding new content, updating plugin.json, validation, and release tagging.

MCP is built on JSON-RPC 2.0 for bidirectional communication between Claude and servers. PROACTIVELY activate for: (1) understanding MCP protocol structure, (2) JSON-RPC message formats, (3) tool vs resource distinction. Triggers: "mcp protocol", "json-rpc", "model context protocol"

Secure MCP servers against prompt injection, tool abuse, excessive permission, and data exfiltration. Covers per-tool scopes, rate limiting, audit logging, and sandbox patterns for shell-adjacent tools. Use this skill when deploying an MCP server to production, handling untrusted agents, or reviewing an MCP server for security issues. Activate when: MCP security, MCP prompt injection, tool sandbox, MCP audit log, MCP rate limit, tool abuse, MCP threat model.

Build Model Context Protocol servers for Claude Code integration

Configure popular MCP servers for enhanced agent capabilities

Memory, context, and persistent knowledge management for the Kortix agent. Covers: kortix-sys-oc-plugin (observations, LTM consolidation, mem_search, mem_save, session_list, session_get), filesystem persistence rules, using .MD files for plans/notes/project state, how filesystem writes feed the memory pipeline, and best practices for ensuring nothing important is ever lost. Load this skill when you need to: understand how your memory works, decide where to persist information, write plans or notes, manage project context across sessions, or optimize your context window usage.

Use this skill when implementing model selection for LLM applications. Activate when the user needs to choose between different AI models, implement cost-efficient model routing, balance quality vs cost, or build intelligent model selection systems.

Monitors AI agent health across quality, cost, performance, and errors. Only use when the user has Amplitude Agent Analytics instrumented in their project. Use when the user asks "how are our AI agents doing", "AI quality check", "agent health", "AI errors", "agent performance", "LLM cost", or wants a proactive health report on their AI/LLM features.

Use when evaluating LLMs, running benchmarks like MMLU/HumanEval/GSM8K, setting up evaluation pipelines, or asking about "NeMo Evaluator", "LLM benchmarking", "model evaluation", "MMLU", "HumanEval", "GSM8K", "benchmark harnesses"

Reference for nimble tasks and batches commands. Load when polling async task status, tracking batch progress, or fetching results. Works for ALL async types: agent run-async, agent run-batch, extract-async, extract-batch, crawl (per-page tasks), search async, map async. CRITICAL: agent tasks use "success"/"error" states; crawl page tasks use "completed"/"failed".

DEPRECATED legacy alias — a2a-pay support has been merged into `okx-agent-payments-protocol` (see `references/a2a_charge.md`). This stub exists only so that legacy invocations referencing the old name resolve to a meaningful redirect. If you land here, immediately load `okx-agent-payments-protocol` via the Skill tool and follow it instead. Do NOT execute any payment logic from this stub. All paymentId / a2a_... / payment-link / payment-status triggers now route directly to `okx-agent-payments-protocol` (Path B in its SKILL.md).

OMC agent catalog, available tools, team pipeline routing, commit protocol, and skills registry. Auto-loads when delegating to agents, using OMC tools, orchestrating teams, making commits, or invoking skills.

Build a retrieval-optimized knowledge layer over agent documentation in dotfiles (.claude, .codex, .cursor, .aider). Use when asked to "optimize docs", "improve agent knowledge", "make docs more efficient", or when documentation has accumulated and retrieval feels inefficient. Generates a manifest mapping task-contexts to knowledge chunks, optimizes information density, and creates compiled artifacts for efficient agent consumption.

Wire encrypted credentials to environment variables using the credential: convention. Use when setting up LLM provider keys (ANTHROPIC_API_KEY, OPENAI_API_KEY) or any env var that should come from encrypted credentials.

Use when the user wants to dispatch parallel Claude Code agents across slot clones of a repo, asks to "fan out", "run N in parallel", "use the slots", or wants to coordinate multiple isolated working copies of the same repo. Explains the slot directory layout, when to fan out vs. stay in primary, and the `gh`-driven workflow that ties slots together.

Internal helper agent. Invoked by orchestrator agents via Task tool. PDF accessibility scan configuration manager. Use to create, edit, validate, or explain .a11y-pdf-config.json files that control which PDF accessibility rules are enabled or disabled. Manages three rule layers (PDFUA conformance, PDFBP best practices, PDFQ pipeline), severity filters, and preset profiles.

Scans lyrics for phrases that may match existing songs using web search and LLM knowledge. Use before release to check for unintentional borrowing.

Trial runner and strategy comparison CLI for the Plaited framework. Use when running evals against agent adapters, computing pass@k metrics, or comparing two eval runs.

Use when creating Claude Code plugins, writing skills, building commands, developing agents, or asking about "plugin development", "create skill", "write command", "build agent", "SKILL.md", "plugin structure", "progressive disclosure"

Postman concepts and MCP tool guidance. Loaded when working with Postman MCP tools to make better decisions about tool selection and workarounds.

Use this agent when you need to address comments on pull requests or code reviews by making the requested changes and reporting back on the resolution. This agent handles the full workflow of understanding the comment, implementing the fix, and providing a clear summary of what was done. <example>Context: A reviewer has left a comment on a pull request asking for a specific change to be made.user: "The reviewer commented that we should add error handling to the payment processing method"assistant: "I'll use the pr-comment-resolver agent to address this comment by implementing the error handling and reporting back"<commentary>Since there's a PR comment that needs to be addressed with code changes, use the pr-comment-resolver agent to handle the implementation and resolution.</commentary></example><example>Context: Multiple code review comments need to be addressed systematically.user: "Can you fix the issues mentioned in the code review? They want better variable names and to extract the validation logic"as...

Reference document describing six pre-flight checks (clean git state, not detached HEAD, gh auth, issue exists and OPEN, remote reachable, duplicate-branch warning) as pure bash exit codes with no LLM calls. Reference only (policy document; consumed by `/flow:start` Phase 0).

Conduct a systematic literature survey. Reads the thesis from .papermill/state.md, searches academic sources, classifies references, identifies gaps, generates BibTeX, and updates the state file. Can launch the surveyor agent for deeper autonomous search.

Comprehensive checklist for deploying RAG systems to production with reliability and scale. Use this skill when preparing RAG for production deployment. Activate when: production RAG, RAG deployment, RAG checklist, RAG scaling, RAG monitoring, production-ready RAG.

Use Claude's prompt caching with 5-minute and 1-hour TTLs to slash costs on repeated context — codebases, system prompts, long documents. Covers cache breakpoints, hit-rate optimization, and the common mistakes that silently disable caching. Use this skill when building apps with repeated large context, optimizing LLM spend, or debugging "why are my cache reads zero?" Activate when: prompt caching, cache_control, cache hit rate, 5 minute cache, 1 hour TTL cache, ephemeral cache, reduce Claude cost.

Use when "writing prompts", "prompt optimization", "few-shot learning", "chain of thought", or asking about "RAG systems", "agent workflows", "LLM integration", "prompt templates"

Detect and neutralize prompt injection attacks in OpenClaw skill content, user inputs, and external data sources. Prevents instruction hijacking and context manipulation.

Interactive QA session where user reports bugs or issues conversationally, and the agent files GitHub issues. Use when user wants to report bugs in their own codebase and create Github issues conversationally

Compares original and optimized prompts by parallel execution in git worktrees. Use when evaluating prompt improvement effects or learning prompt engineering through concrete examples.

Descriptive research type sub-skill. Router-dispatched from research-agent parent on signals like "What is", "What are", "Map", "Document", "Inventory", "Survey the field". Loads the Landscape Map pattern (Pattern B). Not directly model-invocable.

Explanatory research type sub-skill. Router-dispatched from research-agent parent on signals like "Why", "Root cause", "Caused by", "What led to", "Explain why". Loads the Causal Analysis pattern (Pattern C) with 5 Whys, Fishbone, and PECO framework. Not directly model-invocable.

Exploratory research type sub-skill. Router-dispatched from research-agent parent on signals like "How does", "How can", "How might", "explore", "discover", "what factors". Loads the Discovery Report pattern (Pattern A) and SPICE framework guidance. Not directly model-invocable.

Parallel read-only multi-agent review of a current git diff or explicit file scope to find behavioral regressions, security or privacy risks, performance or reliability issues, and contract or test coverage gaps. Use when the user asks for a review swarm, parallel review, diff review, regression review, security review, or wants high-signal issues plus a prioritized fix path without editing files.

Build intent routers for agent task distribution

Run a security assessment using deterministic static analysis tools with LLM-powered triage

Create a scheduled task that can be run on demand or automatically on an interval. Distills the current session into a self-contained reusable prompt with cronExpression or fireAt, then calls the create_scheduled_task tool. Use when the user wants a recurring or one-time task scheduled — e.g. 'remind me every Friday', 'run this hourly', 'schedule a daily inbox check'.

Use this agent when performing security audits, vulnerability assessments, or security reviews of code. Triggers on requests like "security review", "check for vulnerabilities", "OWASP compliance check".

Use when delegating a task to Claude Code CLI with session persistence across turns.

Use when delegating a task to Gemini CLI with session persistence across turns.

Detects your Python environment and guides you through installing plugin dependencies. Use on first-time setup or when MCP server fails to start.

Autonomous AI trading agent for Simul8or, a live market simulator.

Generates Claude Code skills from user descriptions. Use when creating new skills, converting documentation to skills, or scaffolding skill structure.

Create and manage prompt snippets — reusable text blocks referenced inside config variation prompts. Keeps common instructions, personas, and guardrails consistent across multiple configs.

Find files with low test coverage and inspect uncovered lines in a SonarQube project (project key optional when MCP integration already defines the default project)

Automate getSorted.de (Sorted) for freelancer invoicing, expense tracking, and German tax submissions (VAT, ZM, annual returns). Uses real-browser automation via Chrome Beta + agent-browser. Triggers on "create invoice", "sorted invoice", "download invoice", "submit VAT", "tax report", "sorted expenses", "/sorted", or any getSorted.de interaction.

Authoring playbook for building agents that read or write tabular data — Google Sheets, Microsoft Excel, CSV, Airtable, Notion databases, or any spreadsheet. Use this when the user wants an agent that updates rows, reads cells, computes totals, generates reports from sheets, syncs data between spreadsheets, or automates anything involving rows, columns, ranges, or worksheets.

Configure the Claude Code status line with VCS-aware scripts showing git branch, jj change ID, bookmarks, context usage, and costs. Use when setting up a statusline, customizing the status bar, adding git or jj info to the prompt, configuring statusLine in settings.json, or troubleshooting statusline scripts.

Guide for creating AI subagents with isolated context for complex multi-step workflows. Use when users want to create a subagent, specialized agent, verifier, debugger, or orchestrator that requires isolated context and deep specialization. Works with any agent that supports subagent delegation. Triggers on "create subagent", "new agent", "specialized assistant", "create verifier". Do NOT use for Cursor-specific subagents (use cursor-subagent-creator instead).

SurrealMCP -- Model Context Protocol server for SurrealDB. Lets MCP-compatible LLM hosts (Claude Desktop, Cursor, GitHub Copilot in VS Code, Zed, n8n) read and write a SurrealDB instance through a single config entry instead of bespoke per-agent integration code. Part of the surreal-skills collection.

Comprehensive Working Memory update with per-step checklists using swe-wm MCP tools

Complete memory system combining LanceDB auto-recall, Git-Notes structured memory, and file-based workspace search. Use when setting up comprehensive agent memory, when you need persistent context across sessions, or when managing decisions/preferences/tasks with multiple memory backends working together.

The complete web toolkit for your agent. Search the web and get answers in milliseconds. Fetch any URL and get clean markdown back. Send a browser agent to navigate sites, fill forms, and extract structured data. Spin up a headless browser for full programmatic control. Use when you need to search the web, extract/scrape data from websites, handle bot-protected sites, or automate browser tasks using natural language.

Use when starting any session with contextd - introduces core tools for cross-session memory, semantic code search, and error remediation. REQUIRES contextd MCP server.

Capture feedback about wf, util, or vault plugins and file it as a GitHub issue for improvement. Use after any session where the tooling fell short — wrong agent behavior, missing skill coverage, misleading prompts, workflow friction, or convention gaps. Only for generic plugin improvements that would help ANY project, not repo-specific config. Files to JSai23/claude-tooling with the plugin-feedback label.

Comprehensive audit of a Webflow site including pages, CMS collections, health scoring, and actionable insights. Use for site analysis, migration planning, or understanding site structure.

Perform exhaustive code reviews using multi-agent analysis and dynamic skill discovery

Guide la création d'un nouveau skill d'agent IA pour le dépôt foundation-skills. Génère le squelette du SKILL.md et du fichier docs, applique les conventions du dépôt (frontmatter, versionnage, structure). À utiliser quand l'utilisateur demande de créer un nouveau skill, écrire un skill, ajouter un skill ou générer le squelette d'un skill.

Use when the user asks to create, review, audit, migrate, or update an AGENTS.md file for a repository, package, subproject, or monorepo. This skill writes agent-facing repository guidance grounded in project evidence, including setup, build/test commands, code style, testing, security, PR rules, nested AGENTS.md scope, and instruction precedence.

This skill should be used when the user asks about "LLM gateway", "Bankr LLM", "LLM credits", "top up credits", "auto top-up", "llm.bankr.bot", "Claude Code gateway", "OpenClaw setup", "Cursor setup", "OpenCode setup", "LLM models", or any LLM gateway configuration or usage topic.

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

INVOKE THIS SKILL when you need human-in-the-loop approval, custom middleware, or structured output. Covers HumanInTheLoopMiddleware for human approval of dangerous tool calls, creating custom middleware with hooks, Command resume patterns, and structured output with Pydantic/Zod.

Continuous maintenance workflow for plugins that target both Claude Code and Codex. Use when a plugin's commands, skills, or agents have changed and the Codex target needs updating.

This skill should be used when the user asks to "manage teams", "create a team", "create a task", "assign a task", "send a message to an agent", "list teams", "check tasks", "broadcast to team", "read inbox", "update task status", or any team/task/agent/message management operation.

Add or update an LLM eval test for a specific tool or collection. Creates a new eval scenario or updates an existing one to cover new tools. Use when adding eval coverage for new or modified tools.

Build specialized sub-agents for the workflow system

Creates new agent files with complete specifications, following the same format as existing agents. Use when a new agent is needed for tasks that existing agents cannot handle. Generates agent markdown files with name, description, core principles, workflow, and integration details.

Multi-agent orchestration enforcement for /team command. Provides claudish CLI validation, validates session directory paths, and ensures /team internal Tasks use dev:researcher. Use when debugging /team orchestration failures.

Test installed agent guardrail hooks by sending trigger phrases through the stop hook script. Checks hook installation first, then runs two test phrases per rule. Use when user asks to "test guardrails", "verify hooks", "check guardrail patterns", "agent-guardrails test", or wants to confirm hooks are working.

Universal quality bar and final audit rubric for any agent system prompt. Activate this whenever you are unsure which archetype skill applies, or as a final review pass before writing the system prompt. It defines the required run contract, completion criteria, fallback paths, response format, and anti-patterns every produced agent prompt must satisfy.

Guide for orchestrating and controlling Claude Code agent teams. Use when working with or managing multi-agent team coordination in Claude Code, including: (1) Creating and starting agent teams, (2) Controlling teammates (display modes, task assignment, delegation), (3) Best practices for parallel work, (4) Troubleshooting team issues. Covers concepts like team lead, teammates, shared task lists, and inter-agent messaging.

Build autonomous RAG agents that reason, plan, and use tools for complex retrieval tasks. Use this skill when simple retrieve-and-generate isn't enough. Activate when: agentic RAG, RAG agent, multi-step retrieval, tool-using RAG, autonomous retrieval, query decomposition.

Audit and trim AI agent instruction files (AGENTS.md, CLAUDE.md, CONVENTIONS.md, .cursorrules, etc.) by testing which facts an AI agent can discover from code alone. Use when asked to lint, audit, optimize, prune, or trim any agent instruction file in a repository. Removes redundant documentation that wastes context tokens. Supports a dry-run mode that reports findings without modifying files.

Authenticate with AI Universe MCP server for multi-model commands

Check AI CLI usage/quota for Claude Code, OpenAI Codex, Google Gemini CLI, and Z.AI. Use when user asks about remaining quota, usage limits, rate limits, or wants to check how much capacity is left.

USE FOR AI-grounded answers via OpenAI-compatible /chat/completions. Two modes: single-search (fast) or deep research (enable_research=true, thorough multi-search). Streaming/blocking. Citations.

Search scientific papers and retrieve structured experimental data extracted from full-text studies via the BGPT MCP server. Returns 25+ fields per paper including methods, results, sample sizes, quality scores, and conclusions. Use for literature reviews, evidence synthesis, and finding experimental details not available in abstracts alone.