Skills
96,981 packaged capabilities and playbooks for agents. Install any of them with one command.
Agent workflows
You know SEO matters but you don't have time to do daily audits, track rankings, or validate schema markup. This is the AI agent SOP — set it up once, let your coding agent run it daily on autopilot. What's inside: • Daily 4-phase audit automation (technical → content → performance → report) • GSC data pipeline (pull, analyze, prioritize keywords automatically) • AI traffic tracking (ChatGPT/Perplexity referrals in GA4) • Schema validation + auto-fix workflows • IndexNow integration for instant indexing on publish • Week 0-4 ramp-up timeline (from zero to ~32K impressions) Built from: Designed for Claude Code / OpenClaw / Cursor autonomous execution. Validated with real content site data. Triggers: "SEO agent" | "SEO automation" | "SEO Agent SOP" | "autonomous SEO" | "SEO daily report" | "SEO Agent运营" | "SEO自动化" | "GEO agent" | "32K impressions" | "3.2万曝光" | "keyword mapping" | "CTA conversion SOP" | "SEO报告模板" | "IndexNow setup" | "AI crawler robots.txt" | "DataForSEO" | "SEO工具链" | "SEO Agent 日报"
You know you should post on X every day, but writing tweets eats your evenings — and when an AI writes them for you, they sound nothing like you. This skill turns any AI agent into a ghostwriter that actually sounds like you, with a battle-tested SOP that grew a real account +60% in 45 days, fully agent-operated. What's inside: • Persona calibration system — voice guide, expression rules, dead-opener blacklist • Fact-source database (SOURCE-INDEX) — zero fabricated numbers, every stat traceable • Weekly scheduling system with dedup checks and golden time windows • 6 hard safety rules + pre-publish QC (triple-translation test + 5-point checklist) • Tweet-log & weekly report analytics loop • Content methodology from 3,861-tweet dataset analysis (4 archetypes, golden length, visibility principles) • Real case: 1,150 → 1,837 followers (+60%) in 45 days, 1 tweet/day. Triggers: "twitter automation" | "x account growth" | "ai agent twitter" | "ai ghostwriter" | "social media agent" | "tweet writing" | "twitter sop" | "grow twitter followers" | "推特运营" | "AI代写推文" | "X 增长" | "Twitter自動運営" | "AIゴーストライター" | "트위터 자동화" | "AI 고스트라이터"
Set up Claude Code hooks to block dangerous git commands (push, reset --hard, clean, branch -D, etc.) before they execute. Use when user wants to prevent destructive git operations, add git safety hooks, or block git push/reset in Claude Code.
Delegate a bounded task to another coding-agent CLI the way a parent agent delegates to a subagent. CLIs are Claude Code, Codex, Grok Build, Antigravity (agy), and the Cursor CLI (cursor-agent). Use when the user says harness-subagent, or names a CLI or a model such as Astra, Sol, Opus, Opus 5.5, Fable, Sonnet, Sonnet 5.5, Haiku, or Grok 4.7. Do not use to install those CLIs. Do not use if this session was launched by this skill (the prompt starts with YOU ARE THE WORKER, or HARNESS_SUBAGENT_RUN is set). Do that job. Do not delegate again.
Improve existing agent skills based on user feedback and best practices. Use when the user wants to fix, enhance, or refactor an existing skill. Gathers user feedback first, then applies technical analysis and implements improvements.
Interact with InteractiveAI and access its documentation. Use when needing to (1) manage platform resources such as services, secrets, traces, vector stores, and prompts via the `iai` CLI, (2) look up InteractiveAI documentation, concepts, integration guides, or SDK usage, or (3) understand how any InteractiveAI feature works. This skill covers CLI-based platform operations, SDK instrumentation, and documentation retrieval.
Use when a task needs a real iPhone — operating iOS apps that have no API (Apple Health, banking, IM apps), exporting on-phone data, tapping/typing/scrolling on the phone, or taking phone screenshots. Replays saved per-app flows from the iphone-use-flows registry in one call and offers to save new ones after multi-step tasks. Drives the iphone-use daemon through its custom XCTest device runner, HTTP agent API or MCP server.
Access Claude, Gemini, Kimi, GLM and 100+ LLMs via inference.sh CLI using OpenRouter. Models: Claude Opus 4.5, Claude Sonnet 4.5, Claude Haiku 4.5, Gemini 3 Pro, Kimi K2, GLM-4.6, Intellect 3. One API for all models with automatic fallback and cost optimization. Use for: AI assistants, code generation, reasoning, agents, chat, content generation. Triggers: claude api, openrouter, llm api, claude sonnet, claude opus, gemini api, kimi, language model, gpt alternative, anthropic api, ai model api, llm access, chat api, claude alternative, openai alternative
Configure and manage MCP (Model Context Protocol) servers in Claude Code CLI. Use this skill when adding, removing, listing, or troubleshooting MCP servers, or when questions arise about MCP configuration, transport types, scopes, or authentication. Essential for connecting Claude Code to external tools, databases, and APIs.
Manage SSOT, memory, and cross-tool memory search. Guardian of decisions.md and patterns.md. Use when user mentions memory, SSOT, decisions.md, patterns.md, merging, migration, SSOT promotion, sync memory, save learnings, memory search, harness-mem, past decisions, or record this. Do NOT load for: implementation work, reviews, ad-hoc notes, or in-session logging.
Autonomously completes a confirmed milestone by reconciling existing work, coordinating work-item delivery and the configured milestone release, and closing the verified milestone. Use when the user runs /milestone-rush or asks to complete one exact, confirmed milestone, including selecting it after /roadmap-review.
Design n8n AI agents the right way. Use when building or editing any @n8n/n8n-nodes-langchain.* AI node — an AI Agent, LLM chain, Text Classifier, or Information Extractor — and whenever the user mentions AI agents, LLM with tools, tool calling, $fromAI, system prompts, agent memory, sessionId, structured/JSON output, output parser, RAG, vector store, a chat assistant/bot, or human-in-the-loop review. Covers Agent-vs-chain-vs-classifier choice, the model/memory/tools/outputParser slots, tool names/descriptions as prompt, structured output with autoFix, memory, RAG, human review, and chat topologies.
(forwward) Configures a new AI agent through 10 discovery questions, creates platform-specific config files (Claude Code, Codex, Antigravity, Cursor, OpenClaw), tests integrations, and sets security guardrails. Triggers on setting up a new AI agent, configuring a workspace, or initializing agent context.
Coordinate an end-to-end Jira ticket workflow from ticket fetch through per-task implementation. Use this skill when the user provides a Jira URL, says "work on ticket PROJECT-123", "resume PROJECT-123", "continue this Jira ticket", "start the Jira workflow", or asks for status on a ticket without naming a specific phase. This top-level coordinator keeps SKILL.md as a routing layer, loads bundled references just in time, and dispatches execution-heavy work to downstream skills or co-located utility subagents.
Central entry point of the afonsoft agent harness. Use when starting a new project, resuming an existing one, planning features/Epics/releases, or running any multi-step agent-driven work. Validates and reconciles SPECs (SDD), audits the codebase and harness for gaps (security, architecture, performance, hygiene), proposes improvements, fragments work into GitHub Issues, delegates implementation/QA/review to specialized skills, and re-validates everything until delivery. Also use to review unapproved SPECs, reconcile open GitHub Issues with code, or run a final gap check before closing a release.
Diagnose OpenViking memory issues by tracing backward from a memory file to its archive memory_diff.json and, when needed, session messages. Read-only; use for incorrect content, wrong paths or owners, missing memories, and unexplained updates or deletes.
AI-assisted pair programming with multiple modes (driver/navigator/switch), real-time verification, quality monitoring, and comprehensive testing. Supports TDD, debugging, refactoring, and learning sessions. Features automatic role switching, continuous code review, security scanning, and performance optimization with truth-score verification.
Divide-and-conquer implementation from specs/plans. Decomposes a reference document into independent tasks, assigns each to a builder agent, executes in parallel waves respecting dependencies, then integrates results. Use when you have a spec, PRD, plan, or large feature to implement quickly with parallel execution.
Improve router-facing pipeline and reflection narration to reduce noisy status churn and make Step 0/Reflection outcomes explicit. Use when updating Router output contract, reflection reminder wording, or post-pipeline notification batching.
Use when capturing an idea, drafting a design, or tracking work-in-progress in a repo's plans/ directory — creating or updating a plans/YYYY-MM-DD-topic.md file (including a bug, idea or risk turned up incidentally), asking what plans exist or what to work on next, here or across every repo, choosing or advancing a status, retiring a landed/abandoned plan once its content has a permanent home elsewhere, migrating a repo's legacy monolithic plan file (PLAN.md, DESIGN.md, ...) onto this convention, or auditing AGENTS.md/README.md/docs for planning/status/future-work content that has drifted in and belongs in plans/ instead. Also owns where a plan file may live and what may be written in it: a work, client or employer repo that cannot take a plans/ directory keeps its plans in the store outside every working tree ($PLANS_HOME), routed per repo by config; an idea with no repo yet is filed unscoped and graduated later; and no plan committed to a repo you publish may name a client, employer or internal project.
Write, check, repair and publish a plan page: a project's plans and subject files under its plans directory, rendered by .agents/pstack/plan-page.mjs and rendered as one local HTML page per subject with a local index of every subject, and published to claude.ai at each hand-back for comments and feedback. The page is how planned work hands back to the user. Use before writing or changing a plan, a subject file in <plans>/topics or its page; at every stop that hands back work with a plan, such as a review verdict other than a review-only panel's, a next answer or a playbook's close; before writing or updating a PR body, which renders from the PR's plan, or a comment about your commits on someone else's PR; when a plan page or its claude.ai artifact looks wrong or refuses to render; to republish a page; or to change the page shape or a playbook's page sections.
Two agent-readiness scanners in one terminal: every check isitagentready.com and is-agentic.com run, plus copy-paste fixes, CI gating, scan history, and a local store neither web UI has an answer for. Trigger phrases: `scan my site for agent readiness`, `is my site agent-ready`, `what's my agent readiness level`, `what does is-agentic say about my site`, `cross-check my agent readiness score`, `compare both agent readiness scanners`, `how do I make my site ready for AI agents`, `fix my site for AI agents`, `check llms.txt and MCP discovery on this site`, `use isitagentready`, `run isitagentready`.
Master prompt engineering for AI models: LLMs, image generators, video models. Techniques: chain-of-thought, few-shot, system prompts, negative prompts. Models: Claude, GPT-4, Gemini, FLUX, Veo, Stable Diffusion prompting. Use for: better AI outputs, consistent results, complex tasks, optimization. Triggers: prompt engineering, how to prompt, better prompts, prompt tips, prompting guide, llm prompting, image prompt, ai prompting, prompt optimization, prompt template, prompt structure, effective prompts, prompt techniques
Adds new AI providers to claude-council, configures provider API settings, troubleshoots provider connections, and documents the provider script interface. Covers creating provider shell scripts, setting API keys, and validating connectivity. Triggers on "add provider", "new AI agent", "provider not working", "API configuration", or "extend council".
Evaluate how well a codebase supports autonomous AI-assisted development. Analyzes repositories across five pillars (Agent Instructions, Feedback Loops, Workflows & Automation, Policy & Governance, Build & Dev Environment) covering 74 features. Use when users want to assess how agent-ready a repository is.
Use when releasing or upgrading the common-tools, dev-skills, or low-frequency-skill plugins, synchronizing Claude/Cursor/Codex manifests and marketplaces, updating CHANGELOG files, validating README and installation-document consistency, or generating/checking the skill-registry.json discovery manifest for Skill Router MCP. Use when the user requests plugin release, version upgrade, changelog, manifest, marketplace, README consistency, skill-registry.json, MCP skill registry, or registry stale validation.
This skill should be used when the user asks to "fix my skill" or "audit this skill". Make sure to use this skill whenever the user mentions skill quality, structural issues, broken skills, or skill diagnostics — even if they don't explicitly say "repair-skill". Not for adding features or improving effectiveness — use improve-skill. Not for agents — use repair-agent.
This skill should be used when the user asks to "create a plan for replit", "break down tasks", "create development phases", "checkpoint strategy", or needs to convert a project into iterative development phases that Replit Agent can execute step-by-step with checkpoints.
Run a research Workflow end to end when the user requests a survey, brief, review, tutorial, or idea exploration; route an unbound goal, execute one eligible Unit at a time, and stop at checkpoints or diagnosed failures.
Multi-agent code review without any edits or commits — requested rounds of parallel reviewers, synthesis, default-on reproduction of uncertain material findings, and a final report. Use when the user wants a code review of a diff, branch, path, or PR but does NOT want any fixes applied automatically (no edits, no commits, no staging). Activates the `review-anvil` engine in read-only mode.
Routing feature planning through the ring-pm-team pre-dev workflow: choosing the Small Track (4 gates, <2 days) or Large Track (8 gates, 2+ days) and entering via ring:planning-small-features or ring:planning-large-features. Indexes pre-dev gates, standalone utilities, and research agents. Use when starting a feature that needs systematic planning. Skip for quick exploratory work, known-solution bug fixes, or trivial changes.
Using the ring-tw-team plugin and orchestrating its documentation specialists in parallel: guide-writer (guides/concepts/tutorials), api-writer (REST API reference), and docs-reviewer (quality review). Use when creating functional or API documentation, or reviewing doc quality. Skip when writing code (use dev-team) or writing plans (use pm-team).
Writing a rolling-wave phased implementation plan from a spec before coding: a phase-epic-task hierarchy where Phase 1 is detailed into dispatch-ready tasks and later phases stay epic-level for elaboration during execution. Use when a multi-file feature needs decomposition; runs after ring:exploring-codebases or pre-dev gates, hands off to ring:executing-plans or ring:running-dev-cycle. Skip for single-file changes or spikes.
New to MCP?
From what the protocol is to installing your first server, these answers cover most first questions.
An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.