AI Agent Workflow Collection

Senior-PM posture frame plus parallel adversarial review orchestrator. Use this skill at the start of any non-trivial session, OR whenever the user wants a plan, code file, or content piece reviewed before they ship it. Trigger on /captain-opus alone (loads PM posture) or /captain-opus (spawns two reviewers in parallel, synthesizes their disagreements). Especially useful before high-stakes commits, before publishing client-facing content, when a single review felt thin, or when the user says 'make sure this is right' about anything that took meaningful work to produce.

Fetch rendered public web pages via isolated headless Chrome. Default for 'fetch this URL', 'what does this page say', 'scrape this page'. Lighter than Playwright MCP, works in dispatch jobs. Routes authenticated hosts to claude --chrome. Escalates to Playwright MCP only for clicks, form fills, or multi-page flows.

Use when the user says "/clear-prep", "prep for context clear", "prepare to clear", "handover prompt", "I'm going to clear the context", "relay context to a new instance", "snapshot my session", "session is getting full", "give me a handover", or "context is running out". Generates a self-contained handover prompt the user pastes into a cleared Claude Code instance so work continues seamlessly with zero re-explanation.

Execute READY plans from the fleet triage queue. Shows which plans scored ready, asks for confirmation, then dispatches agents. Triggers on: /fleet-execute, 'run fleet', 'execute ready plans', 'dispatch fleet', 'run the fleet'.

Show current work status (active increments, progress) and background jobs (imports, cloning). Even with no jobs, shows increment summary and helpful context.

Have multiple specialized agents review a plan in parallel. Spawns DHH, Kieran, and code simplicity reviewers to provide diverse feedback on implementation plans.

Pre-deploy verification gate. Checks git status, build, tests, migrations, secrets, and commit conventions. Reports READY or NOT READY. Read-only, no changes. Trigger words: preflight, pre-deploy check, ready to deploy, deployment checklist.

Use when the user asks to "spawn an agent team", "run agents in parallel", "multi-agent review", "build an agent team", "fan out work to agents", or asks for parallel execution across multiple workstreams with live synthesis. Coordinates 2 to 10 specialized agents (Sonnet and Haiku by default, Gemini for contrarian or SEO/content, Opus only on explicit ask) using structured templates (3/3/3/1, 2/2/1, or 4/4+1+1), failure recovery, and lead synthesis. Use whenever the work has multiple defensible answers, benefits from independent perspectives, or needs adversarial pressure on a single hypothesis.

Show available skills, plugins, and commands organized by category. Use when starting work, unsure which tool fits, or need a quick reference. Triggers on: 'what tools do I have', 'which skill', 'what commands', '/tool-suggest'.