Skills
97,222 packaged capabilities and playbooks for agents. Install any of them with one command.
Agent workflows
Manage Alibaba Cloud beebot (Chatbot) via OpenAPI/SDK. Use whenever the user asks to configure, query, or troubleshoot Alibaba Cloud chatbot resources, including bot inventory, configuration changes, status checks, and API-level diagnostics.
Skill-set loader for /hunt orchestrator. Fingerprints the target, picks the right platform attack skills, and loads the Red Team or WAPT skill set. Use when /hunt has just received a mode answer (redteam or wapt + blackbox|greybox) and needs to load the appropriate skills and print the taxonomy. Not for direct user invocation.
Manage Claude Code permissions via Dippy config files (global and project-level) and settings.json (for non-Bash/non-MCP permissions like Read, WebFetch, Skill). Use when user asks to "permit", "allow", "whitelist", or "add permission" for a command or tool. Also use when troubleshooting "permission denied" errors, reviewing current permissions, or cleaning up stale entries.
Use this skill when the user asks for efficient communication, task reports, file-operation summaries, research discussion, study-note synthesis, planning, writing feedback, or responses that need conclusion-first structure, concrete evidence, risk disclosure, and useful next steps
This skill should be used for diagnosing and mitigating context degradation: lost-in-middle failures, context poisoning, context clash, context confusion, attention-pattern issues, and agent performance degradation caused by accumulated or conflicting context.
Condenses the current conversation into a handoff document that another agent or session can pick up without prior context: goal, status, next steps, key artifacts by reference, and constraints. Use when deliberately running “/handoffkit” (optionally with a focus argument) to hand off work, write a handoff document, or summarize the session for the next agent.
Creates reusable prompt templates with strict output contracts, style rules, few-shot examples, and do/don't guidelines. Provides system/user prompt files, variable placeholders, output formatting instructions, and quality criteria. Use when building "prompt templates", "LLM prompts", "AI system prompts", or "prompt engineering".
Parallel DAG-plan implementation skill. Reads a v-planning plan directory (root.md + step-<n>.md files), topologically schedules ready steps, and fans them out as parallel sub-agents. Use whenever the user invokes /v-implement, points at a plan directory produced by /v-plan, or asks to "run the parallel plan", "implement the DAG", or "fan out the steps" — even without those exact words. For linear plans (single .md file), use `implementing` instead.
Writes, tests, registers and debugs Claude Code hooks (PreToolUse, PostToolUse, SessionStart, Stop) that turn a rule the model keeps breaking into a hard gate. Use when the user wants to block or intercept a tool call, add a guard rail, fix a misfiring hook, or says 拦截 / 守卫 / 钩子 — including "make it stop doing X".
Execute a single DAG step as an autonomous background sub-agent. Sibling of phase-running for DAG plans produced by v-planning. Reads a step-<n>.md file directly, atomically claims it via frontmatter status, runs the three-bucket Success Criteria, and reports back. Spawned by v-implementing or by /run-step.
Vertical / parallel implementation planning skill. Creates DAG-structured plan directories where each step is an independent, QA-able vertical slice that sub-agents can pick up and implement in parallel. Use whenever the user wants a plan that fans out (multiple independent features), invokes /v-plan, or asks for a "parallel plan", "DAG plan", "vertical plan", or "plan that can be parallelized" — even if they don't say those exact words. Prefer the linear `planning` skill for strictly sequential work.
Discover and reconcile Alibaba Cloud product catalogs from Ticket System, Support & Service, and BSS OpenAPI; fetch OpenAPI product/version/API metadata; and summarize API coverage to plan new skills. Use when you need a complete product list, product-to-API mapping, or coverage/gap reports for skill generation.
Eval and improve a skill against golden cases — run the target skill blind in a fresh, context-free subagent on each example input, grade the artifact against the expected outcome, and let the gaps drive the edits. Use when the user wants to test/eval/improve/harden a skill, says "this skill keeps producing X / keeps missing Y", or hands a skill plus example input→expected-output pairs. Pairs with [write-skills](../write-skills/SKILL.md) (the authoring principles every fix obeys).
Use when asked to compare, cross-review, merge, judge, choose, or arbitrate competing plans from multiple agents such as Codex and Claude Code; when given two or more proposed plans, session IDs, transcripts, plan documents, PR descriptions, or pasted strategies; or when the user wants one recommended execution plan after agents review each other's proposals.
Creates, updates, lints, and publishes portable agent skills for Claude Code, Codex, and OpenCode, including repo-local .agents/skills sources, Claude compatibility symlinks, cross-agent SessionStart refresh hooks, and npx distribution. Use when (1) writing or editing a skill, (2) linting with agnix, (3) publishing changes, (4) renaming or deleting a skill, or (5) deciding skill layout, scope, frontmatter, discovery, or refresh automation. Triggers include "new skill", "update skill", "repo skill", "skill not loading", "npx skills", "skills update", "Codex skill hook", "lint skill", "agnix", "SKILL.md".
Designs multi-step agent workflows with tool usage, retry logic, state management, and budget controls. Provides orchestration diagrams, tool execution order, fallback strategies, and cost limits. Use for "AI agents", "agentic workflows", "multi-step AI", or "autonomous systems".
Turns a task the owner keeps doing by hand into a named, reusable skill they can trigger by name or put on a schedule. Watches them do it once or listens to them describe it, works out what varies and what stays fixed, writes the skill with approval checkpoints in the right places, registers it with the router, and shares it with the team. This is how a business gets a workflow nobody shipped for them. Use this whenever the owner describes something repetitive they wish were automatic — including phrasings like "I do this every week," "can you remember how to do this," "make this a thing I can just ask for," "build me an agent," "automate this for me," "turn this into a workflow," or "I want this to happen every Monday." Reach for it when they describe a routine, even without asking for automation.
Route explicit ClawTeam runtime requests into one honest operator packet before touching commands: manual-team, template-launch, monitor-recover, or profile-setup. Use when the user specifically wants ClawTeam's team/task/inbox/worktree/board workflow for parallel developer work, product/ops research, content operations, or game-production coordination. Not for generic multi-agent routing, board-only governance, or lightweight subtask fan-out when jeo, harness, task-planning, or built-in delegation already fit
Typed knowledge graph for structured agent memory and composable skills. Use when creating/querying entities (Person, Project, Task, Event, Document), linking related objects, enforcing constraints, planning multi-step actions as graph transformations, or when skills need to share state. Trigger on "remember", "what do I know about", "link X to Y", "show dependencies", entity CRUD, or cross-skill data access.
Protracted strategy: For long-term, complex tasks that cannot be won quickly, objectively assess the situation, divide the effort into defense / stalemate / counter-offensive stages, determine the current stage, set transition conditions, and identify local offensive opportunities even during difficult stages. Trigger when a task spans multiple sessions or iterations, resources are temporarily insufficient, or either “wanting to give up” or “going all in” arises; do not trigger when the task can be completed in the current session.
Browser automation CLI for AI agents. Use when the user needs to interact with websites, including navigating pages, filling forms, clicking buttons, taking screenshots, extracting data, testing web apps, or automating any browser task. Triggers include requests to “open a website”, “fill out a form”, “click a button”, “take a screenshot”, “scrape data from a page”, “test this web app”, “login to a site”, “automate browser actions”, or any task requiring programmatic web interaction.
Criticism and self-criticism: honestly, specifically, and factually review the work and process after completion, at a review checkpoint, when receiving criticism, or when the same mistake recurs; produce actionable improvements and address external criticism without defensiveness. Trigger signals include review, retrospective, audit, "帮我看看有没有问题", "你确定吗", "check this for problems", and "are you sure". Do not trigger at the start of a task or for a one-step lookup.
Mass line: Collect scattered input from everyone who holds ground truth (users, operators, codebase, logs, runtime results), synthesize it into a coherent plan, take it back for alignment and validation, and iterate. Trigger when inputs conflict, feedback must be consolidated, a plan needs sign-off from real users, or you sense you are designing in isolation; skip when there is a single complete source and the work is pure execution.
Use the public `orca` CLI to operate Orca-managed worktrees, folder contexts, terminals, repos, automations, artifacts, worktree comments, and the browser embedded inside the Orca app. Use when the user says "$orca-cli", "use orca cli", "Orca worktree", "child worktree", "cardStatus", "spawn codex/claude in a worktree", "read/wait/send Orca terminal", "terminal send", "full handoff", "handover", "give this to another agent", "another worktree", "Orca browser", "orca artifacts", "share HTML/Markdown", "public artifact link", or "control the browser inside Orca". Prefer this over raw `git worktree`, ad hoc PTYs, Playwright, or Computer Use when the task touches Orca-managed state. Use Computer Use for browser windows, webviews, or desktop UI outside Orca's embedded browser.
Activate when the user asks about, uses, designs, troubleshoots, or upgrades LiteFlow. Covers LiteFlow 2.16.2 integration and configuration, components, EL, context, executors, scripts, rule sources, Rule-DB, monitoring, dynamic building, testing, source code, Jev smart selection, boolean evaluation and SWITCH/IF routing, as well as AgentScope 2-based liteflow-agent, models, tools, MCP, sessions, events, HITL, Skills, Harness, and A2A. Also helps identify and migrate legacy liteflow-react-agent / ReActAgentComponent usage.
Use when creating, auditing, packaging or publishing an agent skill or Claude Code plugin. Triggers - "make a skill" / "сделай скилл", "skill audit" / «аудит скилов», "заверни в плагин", "опубликуй скилл". NOT for a repo that ships no skill; opt out with "no make-skill" / «без make-skill».
Create a new AI agent skill from scratch (kit-convention naming, drafting, trigger cases, live testing, publishing), or re-check an existing one after a change. Use when the user says "help me make a skill for X", runs "/skillkit", or says "I changed this skill, check it".
Run a groomed `ready` GitHub issue through the entire build cycle unattended: worktree, implementation, commit, QA plan, acceptance check, fixes, and PR creation. Takes one issue to a reviewable PR without human intervention. Use when the user says "run issue #N unattended" or "work the ready issues while I'm away". Starts from a groomed issue; planning and grooming remain attended
Foundry IQ knowledge bases. WHEN: make local or Blob documents searchable; create/diagnose KBs; triage unsupported connectors or multi-source KB creation/reconfiguration (even with classic-search terms); connect existing KB to agents (including multi-source KBs); create/reuse a Search service; retrieve from an existing knowledge base with citations; no brand words needed. Read its procedure before query/target questions. NOT: other KB providers, repository-file search, classic Azure AI Search index/query/app work with no Foundry IQ KB or connector request, generic agent creation.
Persistent Obsidian-based memory for coding agents. Use at session start to orient from a knowledge vault, during work to look up architecture/component/pattern notes, and when discoveries are made to write them back. Activate when the user mentions obsidian memory, obsidian vault, obsidian notes, or /obs commands. Provides commands: init, analyze, recap, project, note, todo, lookup, relate.
Analyzes and improves prompts using 31 frameworks across 7 intent categories. Use when a user wants to improve, rewrite, structure, or engineer a prompt—including requests like “help me write a better prompt,” “improve this prompt,” “what framework should I use,” “make this prompt more effective,” or any prompt engineering task. Recommends the right framework based on intent (create, transform, reason, critique, recover, clarify, agentic), asks targeted questions, and delivers a structured, high-quality result
This skill should be used for the tool-interface layer of an agent system specifically: writing tool descriptions agents can route on, designing tool schemas and response formats, naming conventions, actionable error recovery messages, MCP server design, tool-set consolidation, and deciding when to add or remove an individual tool. Use this when the unit of work is a single tool or a set of tools. Route project-shape, pipeline architecture, and task-model-fit decisions to project-development; route deciding whether to introduce sub-agents to multi-agent-patterns.
Create, improve, or audit Vapi voice agent and Squad system prompts for production phone and web based voice agents. Use when the user wants help designing a Vapi assistant prompt, multi-assistant Squad prompt set, refining an existing prompt, creating prompt sections, building an intake or handoff workflow, improving tool-use instructions, adding guardrails, or optimizing voice-agent behavior for brevity, turn-taking, error handling, caller data collection, escalation, handoffs, and spoken formatting.
Compress large context before reasoning to reduce token usage while preserving evidence. Use this whenever the user mentions huge files, long prompts, RAG payloads, prompt caching, expensive sessions, codebase context, chat history compaction, or wants the same answer quality with fewer tokens.
Build, modify, debug, and deploy agents with Agentforce Agent Script. TRIGGER when: user creates, modifies, or asks about .agent files or aiAuthoringBundle metadata; changes agent behavior, responses, or conversation logic; designs agent actions, tools, subagents, or flow control; writes or reviews an Agent Spec; previews, debugs, deploys, publishes, or tests agents; uses Agent Script CLI commands (sf agent generate/preview/publish/test). DO NOT TRIGGER when: Apex development, Flow building, Prompt Template authoring, Experience Cloud configuration, or general Salesforce CLI tasks unrelated to Agent Script
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.