opsMachine/OM-Agency

grumpy-dev-code-review

Performs comprehensive, brutally honest full-codebase review as a grumpy but fair senior developer with 20+ years of experience. Use when the user asks for a grumpy dev code review, brutal code review, senior dev review, comprehensive codebase audit,…

原文语言：英语

onecli

Diagnose and fix OneCLI gateway issues — Docker not running, 407 proxy auth, 401 from API (MITM not active), stale aoc_ token, CA cert refresh, secret setup, version pinning. Use when onecliStart fails, update-fathom-db fails, proxy returns 407 or 401,…

原文语言：英语

ops-machine-crm

ANY CRM or progress.md write, log call, add contact, pipeline lookup, paste transcript/email. Runs lifecycle routing in Step 0 — delivery work goes to progress.md not Won CRM Next Action. Query before create. For week view use week-scan not this skill.…

原文语言：英语

advisor-panel

Convenes a short virtual panel of 2–3 named advisers (verbatim bios below) for questions that merit deeper examination—strategy, ethics, shipping tradeoffs, growth, or inner alignment. Use when the user asks for a board of advisers, adviser panel, stress-test…

原文语言：英语

comedy-db

Manages Mitch's Notion comedy writing database. Use this skill whenever Mitch wants to interact with his comedy material — creating new bit entries, updating existing bits as material develops, reading what's in the database, changing status, or adding notes.…

原文语言：英语

linkedin-comment-engine

LinkedIn organic lead generation system based on the "comment keyword + DM" mechanic. Use this skill whenever Mitch wants to: write a LinkedIn post designed to generate leads, build a resource (checklist, template, list) to use as a CTA, research what topics…

原文语言：英语

mental-model-recipes

Applies the Cognitive Operations Architecture — structured thinking recipes built from lenses (what to notice), operations (what to do), and recipes (specific sequences for specific problems) — to user queries. This skill triggers on ALL user queries unless…

原文语言：英语

skill-writer

Guide users through creating Agent Skills for Claude Code. Use when the user wants to create, write, author, or design a new Skill, or needs help with SKILL.md files, frontmatter, or skill structure.

原文语言：英语

excalidraw-pyramid-parser

Parse and iteratively refine .excalidraw strategic pyramid maps. Use whenever the user uploads a .excalidraw file, mentions a pyramid map, leverage map, strategy canvas, or asks "help me think through this", "what's missing", "where's the leverage". This is a…

原文语言：英语

review-audit

Validate a code review against the actual codebase. Verifies every claim, re-rates severity, finds blind spots, and produces a prioritized action matrix. Invoke with '/review-audit' or 'audit this review', 'verify this code review', 'check if this review is…

原文语言：英语

workflow-router

Workflow manager that orchestrates the entire skill system. Runs automatically before any implementation work. Reads state from artifacts, determines the next skill, spawns sub-agents for execution, and manages human gates. Invoke with '/workflow-router' or…

原文语言：英语

implement-direct

Implement a feature directly from spec without TDD. Use when tests aren't worth the overhead. Invoke with '/implement-direct path/to/spec.md' or 'just implement', 'implement directly', 'skip tests'.

原文语言：英语

implement-to-pass

Implement minimum code to pass a failing test. Use after /write-failing-test confirms red phase. Invoke with '/implement-to-pass' or 'green phase', 'make test pass', 'implement the test'.

原文语言：英语

spec-review

Review a spec document for completeness before approval. Use after interview, before implementation, or when asked to 'review spec', 'check spec', or 'is this spec ready'. Read-only analysis that flags gaps.

原文语言：英语

1-security-audit

Phase 1 of security audit pipeline. Scans the codebase for vulnerabilities and creates SECURITY_PLAN.md. Use after /full-security-audit or invoke directly to start a security review. Say 'security audit phase 1' or run '/1-security-audit'.

原文语言：英语

2-security-critique

Phase 2 of security audit pipeline. Red team review of Phase 1 findings — removes false positives, adds missed risks, ranks the backlog. Invoke with '/2-security-critique' after Phase 1 is complete.

原文语言：英语

3-security-spec

Phase 3 of security audit pipeline. Writes a failing test that reproduces the top vulnerability from the ranked backlog. Invoke with '/3-security-spec' after Phase 2 is complete. Do NOT fix code — just write the test.

原文语言：英语

4-security-fix

Phase 4 of security audit pipeline. Implements the fix to pass the failing security test from Phase 3. Loops back to Phase 3 for the next item. Invoke with '/4-security-fix' after Phase 3 creates the failing test.

原文语言：英语

interview

Structured requirements gathering for new features or changes. Use when starting work on a feature, beginning a new task, or when the user says 'interview', 'start work on', 'new feature', or 'requirements'. Reads GitHub issue for context, outputs spec to…

原文语言：英语

diagnose

Bug investigation sub-agent. Receives triage context from the workflow manager, investigates codebase for root cause, and generates a draft bug spec. Dispatched by the manager after triage — do not invoke directly.

原文语言：英语

full-security-audit

Run a comprehensive 4-phase security audit on the codebase. Use for periodic security reviews, before major releases, or when touching auth/payments. Orchestrates the 4-phase security pipeline. Say 'security audit' or 'run security pipeline'.

原文语言：英语

plan-tests

Create a test plan for an approved spec. Use after spec approval, before writing tests. Invoke with '/plan-tests path/to/spec.md' or 'plan tests', 'test planning', 'what tests do we need'.

原文语言：英语

qa-handoff

Update GitHub issue with testing checklist for QA testers. Use after implementation is complete, before merge. Invoke with '/qa-handoff #42 path/to/spec.md' or 'qa handoff', 'ready for testing', 'update issue for QA'.

原文语言：英语

remember

Store a fact, decision, or instruction into the right context file. Project-scoped facts go to .claude/primitives/. Global facts (coding rules, workflow patterns, skill knowledge) go to ~/.claude/ files. Use when the user says 'remember', 'note that', 'add to…

原文语言：英语

scaffold-project

Bootstrap or review project-level context primitives for Claude Code. First run creates .claude/primitives/ and .claude/skills/project-context/. Re-runs review existing primitives against the codebase, auto-update what changed, and flag what's stale. Invoke…

原文语言：英语

supabase-security

Supabase security best practices and patterns. Use when working with Supabase projects, creating tables, writing RLS policies, edge functions, or reviewing Supabase code. Invoke with '/supabase-security' or when asked about Supabase security.

原文语言：英语

webapp-testing

Toolkit for interacting with and testing local web applications using Playwright. Supports verifying frontend functionality, debugging UI behavior, capturing browser screenshots, and viewing browser logs.

原文语言：英语

write-failing-test

Write failing tests for all planned acceptance criteria from the test plan. Use after /plan-tests, before implementation. Invoke with '/write-failing-test path/to/spec.md' or 'write failing test', 'red phase', 'start TDD'.

原文语言：英语