Agent Workflow Skills

Запуск нескольких субагентов параллельно с автоматической сборкой результатов в единый отчёт. Используй для независимых задач которые можно выполнить одновременно. Триггеры: 'параллельно', 'несколько субагентов', 'запусти агентов', 'исследуй рынок', 'проанализируй конкурентов'.

Автоматический запуск и управление субагентами для параллельной обработки задач. Разбивает большие задачи на подзадачи, запускает субагенты параллельно, собирает результаты.

Create a plan for the specified task.

> (1) End-to-end AI generation via Gemini/ZenMux — given a prompt, directly output an image. (2) Code-based rendering — generate HTML/React single-file, render to PNG via Playwright. (3) Prompt engineering — generate optimized prompts for external models (nano-banana-pro, etc.).

Use when the user wants to create a new CLAUDE.md from scratch for a repository, bootstrap project memory for a new repo, set up a repo for Claude Code for the first time, or completely rewrite an existing CLAUDE.md deemed unsalvageable. Produces a lean, high-signal file under 80 lines (120 hard cap) by scanning the repo for mechanical content (stack, commands, structure) and interviewing the user for judgment content (WHY, gotchas, out-of-scope, approval gates, testing philosophy, external docs like Confluence pages). NOT for auditing or incrementally improving an existing CLAUDE.md — that is the claude-md-improver skill's role.

Hand off the current thread to a new Pulumi Neo task as a one-way transfer. Use when the user explicitly asks to hand off, send, transfer, or continue current work in Pulumi Neo (e.g. "hand this to Neo", "continue in Neo", "/neo-handoff"). Do not load when the user only mentions Neo, asks what Neo can do, asks for an AI-written PR or preview explanation, or hands off to a different agent.

> System Architect, Software Architect, Data Architect, and an Adversarial Architect that red-teams failure paths, LLM adversarial output scenarios, silent corruption risks, and invariant bypasses. Each subagent independently analyzes the codebase from their domain perspective, then their findings are consolidated into a living `architecture.md` file with a tracked to-do list and change log. Use this skill whenever the user asks to review, analyze, audit, or improve the architecture — or when they ask questions like "what should we change structurally?", "how is the system organized?", "what are our architecture problems?", "can you do an architecture review?", or "let's redesign X". Also trigger when the user mentions technical debt, structural improvements, security concerns, failure modes, vulnerabilities, or wants a second opinion on design decisions, even if they don't use the word "architecture".

Run mykg knowledge-graph commands inside Claude Code from one slash command `/mykg`. The user describes intent in natural language (extract, append, resume, approve, walkthrough, parse-docs, fetch-web, query); the skill parses intent, builds the right `mykg` CLI command from the live `--help` output, confirms, runs it, and drives the inbox/outbox watch loop for LLM-bearing commands (extract-graph). For read-only queries, prefers MCP tools when the mykg MCP server is online, falling back to reading session files directly, with the `mykg query` CLI as a last-resort fallback. Ensures `.mcp.json` is configured (with user approval). Excludes `mykg init` (interactive shell command) and `mykg merge-graphs` (follow-up planning).

The meta-skill: create AI agent teams, individual agents, or custom skills for any project. Use when the user wants to generate a complete .claude/ agent team, create a single agent, or create a single skill.

Analyze coding sessions to detect corrections and preferences, then propose targeted improvements to Skills used in the session. Use this skill when the user asks to "learn from this session", "update skills", or "remember this pattern". Extracts durable preferences and codifies them into the appropriate skill files.

AI session compression techniques for managing multi-turn conversations efficiently through summarization, embedding-based retrieval, and intelligent context management.

Use multiple Claude agents to investigate and fix independent problems concurrently

Reliably drive the GitHub gh CLI for issue, PR, and label operations in automation and subagent environments, with pre-flight verification so you never fabricate a success or a fake issue URL.

Guide for creating effective skills

Reference implementation demonstrating the Command → Agent → Skill orchestration pattern in Claude MPM, showing both preloaded-skill and dynamic-skill-invocation styles

> Use this skill whenever a new client signs or client onboarding comes up in any up", "take over from their old IT guy", or names a new client that needs setting up. Also trigger for planning or quoting onboarding work, building a credential collection list, deploying agents to a new environment, or the 30-day review. This is the handoff target for msp-sales pipeline stage 7 (Closed Won). Apply alongside msp-legal (paper gates before work starts), msp-pricing (onboarding is always billed separately), msp-brand (every client-facing message), and hand off to msp-helpdesk and msp-qbr at the end.

Analyzes security risks in the AI Agent/MCP Skill catalog. Used when users request to inspect, audit, review, or scan the Skill catalog for potential security risks, including command injection, data leakage, prompt word attacks, stealth access, remote execution, or other malicious activities within the SKILL package.

Use when the user asks Codex, Claude Code, ChatGPT-style coding agents, or another AI agent to build, inspect, run, debug, or revise COMSOL Multiphysics / COMSOL Desktop models. Choose the simplest real COMSOL control path for the task: saved `.mph` inspection, local COMSOL documentation, direct COMSOL executables (`comsolbatch`, `comsolcompile`, `comsolmphserver`, `comsol.exe mphclient`), or the sim COMSOL runtime when structured live introspection, shared Desktop collaboration, checkpointing, or plugin diagnostics are useful. Do not use for generic COMSOL theory.

Build voice AI agents with LiveKit Cloud and the Agents SDK. Use when the user asks to "build a voice agent", "create a LiveKit agent", "add voice AI", "implement handoffs", "structure agent workflows", or is working with LiveKit Agents SDK. Provides opinionated guidance for the recommended path: LiveKit Cloud + LiveKit Inference. REQUIRES writing tests for all implementations.

Deliver a large implementation plan, architecture migration, refactor, or feature roadmap as right-sized, coherent subtasks that are meaningful to code, review, test, and recover independently. Use only when the user explicitly requests this skill or workflow, or explicitly accepts it after an optional proposal during a user-initiated brainstorming or planning session before coding. Do not use for ordinary coding requests with clear requirements and no prior user-requested brainstorming or discussion. This skill does not authorize subagents; before any spawn, explain the usage impact and obtain explicit user approval for the proposed agent count and scope.

Use this skill when reviewing, designing, or improving a workshop, lesson, training session, facilitation plan, talk, or educational explanation. Use it when the user provides a topic, concept, or explanation and wants better framing questions, curiosity hooks, opening prompts, discussion questions, learner reflection prompts, or ways to make participants think before being taught.

Administer the backlog tooling ecosystem when a capability gap is discovered. Invoke when backlog.py, backlog skills, or backlog agents lack a needed operation and a workaround was used or is about to be used. Classifies gaps as script (delegates to @python-cli-architect), process (loads improve-processes), or documentation (delegates to @contextual-ai-documentation-optimizer). Scope — backlog.py, create/work/groom-backlog-item skills, backlog-item-groomer agent, hooks, templates, references, rules, and tests.

Explains Chain of Verification (CoVe) prompt design — a 4-step pattern separating generation from independent factual verification. Use when designing prompts that require factual accuracy, reducing hallucinations, checking technical standards or APIs, or producing step-by-step procedures where subtle errors are costly.

Create GitHub Releases with AI-analyzed changelogs for every calendar day with commits on origin/main. Use when creating daily release notes, backfilling releases, or generating AI-categorized changelogs per day. Uses collect → bucket → analyze → synthesize → publish pipeline via Haiku subagents. Idempotent — skips up-to-date days, updates releases where new commits were added. Accepts optional --start-date, --end-date, --branch, and --dry-run arguments.

Delegation prompt template enforcing WHERE-WHAT-WHY structure for sub-agent prompts. Use when assigning work to a sub-agent, before invoking the Agent tool, preparing prompts for specialized agents, or needing the OBSERVATIONS-SUCCESS-CONTEXT format with authoring rules and pre-send checklist. For comprehensive delegation guidance, activate the agent-orchestration how-to-delegate skill.

Test harness for Claude Code skill argument substitution — demonstrates capture-block pre-declaration, XML tag referencing, unintentional variable corruption in code blocks, and correct placement of shell examples in reference files. Use when verifying substitution behavior before applying a pattern to other skills, testing how arguments flow from skill invocations, or understanding the pre-declaration and reference file pattern with greet/farewell/inspect actions.

Integrates patterns from external sources (URLs or local files) into local skills, agents, and plugins. Use when comparing external agent definitions against local equivalents, extracting best practices from frameworks like GSD or BMAD-METHOD, enhancing local skills with external workflow patterns, or adding interoperability with external tool ecosystems. Runs a 3-phase workflow — parallel candidate mapping, contextual enhancement, and validation — with source attribution and backlog tracking for deferred enhancements.

Manages the research/ knowledge base of verified tool and library research entries via knowledge-explorer.py. Commands — list (browse by category or SDLC layer), fetch-github (draft entry from GitHub repo), add (validate and route to category dir), update-append (append dated revision), generate-descriptions (repair missing or low-quality descriptions), migrate (convert inline-header entries to frontmatter). Use when browsing KB topics, adding or updating research entries, fetching GitHub repo metadata, generating descriptions in parallel via Haiku subagents, or migrating old-format entries to skill-spec frontmatter.

Produces a structured, end-to-end linear walkthrough of an unfamiliar codebase by orchestrating parallel subagents across four phases — discovery, tracing, validation, and synthesis. Use when onboarding to a new repository, understanding execution paths from entry points, generating navigable codebase documentation, or needing fact-checked coverage of architecture, deployment, testing, and operations.

Facade skill for multi-agent swarm orchestration in Claude Code — routes to specialist skills covering primitives, spawning, operations, and patterns. Use when coordinating multiple agents in parallel, building pipeline workflows with dependencies, running parallel code reviews, creating self-organizing task queues, designing divide-and-conquer workloads, or choosing between TeamCreate and Agent tool approaches. Loads swarm-primitives (team lifecycle and message flow), swarm-spawning (agent types and backends), swarm-operations (tool API and error handling), and swarm-patterns (six orchestration recipes with complete workflows).

Transforms validated /linear-walkthrough artifacts into one presentation-ready deck outline per major codebase component. Use when the user asks to create a presentation, prepare a walkthrough deck, build an onboarding deck, produce an architecture review deck, or summarize a codebase for a technical audience. Requires walkthrough directory output -- reads unified walkthrough, per-section files, validation reports, coverage maps, entry points, and open questions -- then orchestrates four parallel agent phases to produce slide outlines with speaker notes, evidence references, and suggested visuals.

Look up prior Claude Code sessions when context is lost or forgotten. Use when asked what was done before, what happened in the last session, or any request to recall past conversation history, prior decisions, experiments, or outcomes. Indexes and searches raw JSONL transcripts from ~/.claude/projects/ via DuckDB. Returns verbatim user messages, summarizes AI actions and sub-agent outcomes, detects tool errors and user frustration signals, and reports tool usage statistics. Summaries cached at ~/.claude/kaizen/session-summaries/.

Walks a user from one rough content idea to a finished post, platform-adapted variants, and repurposed follow-on assets using a 7-step sequential prompting workflow. Use when the user wants to turn a half-formed idea, frustration, story, or client situation into publishable content quickly and consistently — covering idea extraction, hook generation, outline, draft, humanizer pass, platform adaptation (LinkedIn, Reddit, X/Twitter, newsletter, Telegram), and repurpose engine.

Builds comprehensive Claude Code skills using parallel research agents — categorization, parallel documentation gathering, anti-hallucination checkpoints, and final validation. Use when building a skill from official docs, when "research for skill" or "create comprehensive skill" is requested, or when extensive multi-source documentation gathering is needed before skill creation.

API reference for Claude Code multi-agent swarm tools -- TeamCreate, SendMessage, TeamDelete, TaskCreate/Update/List/Get, and Agent tool parameters. Use when looking up tool signatures, message schemas, shutdown sequences, error handling patterns, or debugging swarm operations. Covers direct messages, broadcasts, plan approval flows, graceful shutdown sequences, crashed teammate recovery, and common error causes.

Covers how to create agents in Claude Code swarms -- subagents vs teammates, built-in agent type catalog (Bash, Explore, Plan, general-purpose, claude-code-guide), plugin agent types from compound-engineering, spawn backend selection (in-process, tmux, iterm2) with auto-detection logic, and environment variables injected into spawned agents. Use when choosing how to spawn an agent, selecting the right agent type, configuring or troubleshooting spawn backends, or passing env vars to teammates.

Scientific delegation framework for orchestrators coordinating sub-agents. Provides WHERE-WHAT-WHY context patterns while preserving agent autonomy. Use when delegating tasks, structuring sub-agent prompts, planning multi-agent workflows, or coordinating specialist agents.

Quick delegation template for sub-agent prompts. Use when assigning work to a sub-agent, before invoking the Agent tool, or when preparing prompts for specialized agents. Provides the WHERE-WHAT-WHY framework. For comprehensive delegation guidance, activate the agent-orchestration how-to-delegate skill.

Scientific delegation framework for orchestrators — provide observations and success criteria while preserving agent autonomy. Use when assigning work to sub-agents, before invoking the Agent tool, or when preparing delegation prompts for specialist agents.

Master multi-agent orchestration using Claude Code's swarm system. Use when coordinating multiple agents, running parallel code reviews, creating pipeline workflows with dependencies, building self-organizing task queues, or any task benefiting from divide-and-conquer patterns. This facade loads specialist skills for primitives, spawning, operations, and patterns.

Transform transcript analysis findings into actionable improvements. Triggers on "generate hooks from findings", "improve agent", "fix anti-pattern", "kaizen improvement", "generate hook proposals", or "create improvement plan". Provides templates for hook generation, agent prompt refinement, skill patches, CLAUDE.md updates, and script automation from analysis data.

Use when extracting specific data points from large agent output transcripts, kaizen analysis reports, or JSONL session files — tool timings, query counts, error summaries, or any structured facts — without loading raw data into orchestrator context. Activates when the orchestrator needs targeted facts from large files and context pollution must be avoided.

This skill should be used when analyzing Claude Code session transcripts, reviewing agent performance, finding anti-patterns or tool misuse, detecting user frustration signals, mining workflow patterns, running kaizen analysis, debugging agent behavior, or performing session forensics. Provides JSONL schema (kaizen-analysis get_transcript_jsonl_schema or MCP resource kaizen://session-log/schema or references/jsonl-schema.md), arbitrary DuckDB SQL over JSONL via kaizen-duckdb execute_query, cookbook query patterns, 10 analysis dimensions, and PM4Py process mining methodology.

Administer the backlog tooling ecosystem when a capability gap is discovered. Invoke when backlog.py, backlog skills, or backlog agents lack a needed operation and a workaround was used or is about to be used. Classifies gaps as script (delegates to @python-cli-architect), process (loads improve-processes), or documentation (delegates to @ai-doc-optimizer). Domain: backlog.py, create/work/groom-backlog-item skills, backlog-item-groomer agent, hooks, templates, references, rules, and tests.

Use when orchestration or planning agents are producing task plans, task prompts, or TASK.md instructions that must be unambiguous, verifiable, and resistant to hallucination. Applies CLEAR (Concise, Logical, Explicit, Adaptive, Reflective) to structure and write agent task files, then adds CoVe (Chain of Verification) checks where accuracy risk is meaningful. Activates on draft task prompts, swarm plans, migration tasks, and multi-step plans requiring independently executable steps.

Loaded automatically when reviewing Claude skills or agent definitions — covers SKILL.md structure, frontmatter validity, token budget, description quality, and agent contract compliance.

Orchestrate parallel agent teams as a manager — not a micromanager. Use when coordinating 2+ independent workers, running SAM task waves, relaying discoveries between worker waves, handling blockers, or synthesizing team results. Covers both SAM structured dispatch (task file does the work) and ad-hoc dispatch (reference agent-orchestration for prompt template).

Executes SAM Stage 5 — dispatches a single ARTIFACT:TASK file to a fresh stateless agent session, runs quality gates, and produces an ARTIFACT:EXECUTION with implementation results and verification output. Use when Stage 4 Task Decomposition is complete and tasks are ready for execution, when re-executing a task after Stage 6 returns NEEDS_WORK, or when dispatching a task to a language-appropriate specialist agent via the development harness pipeline.