AI Agent Workflow Collection

Convene a council, ask the council, debate this, get a panel to critique, run this by a council of experts, council of advisors, devil's advocate panel, red-team this decision, what would a panel say, multi-perspective critique, stress-test this decision, get diverse expert opinions and a recommendation. Convenes 3–5 cognitive-framework advisors that debate any decision — repo plans, website/UX designs, product or codebase critiques, marketing plans, strategy choices — then vote and return one ranked, confidence-rated recommendation with dissent preserved and a single concrete next step. Use when a decision genuinely merits several opposed expert lenses; not for quick single-answer questions or line-by-line code review (use rad-code-review for that).

This skill should be used for a quick health/status overview of a Coolify instance — a one-shot "is everything up" dashboard. Trigger when: "coolify status", "is everything up", "how's my Coolify doing", "status check", "give me a Coolify overview", "anything down?", "morning check", "is my server healthy", "what's the state of my deployments". For deep diagnosis of a specific problem, use coolify-troubleshoot; for performing actions (deploy, restart, env vars), use coolify-actions.

Use when the user says "draft a pitch", "write a pitch for [project]", "one-pager", "executive summary", "investor pitch", "grant proposal", "pitch deck content", "partnership memo", "draft a funding pitch", "write a long-form pitch", or wants a longer-form pitch document (one-pager, executive summary, or deck-equivalent text) for a project. Different shape from elevator-pitch (which is the ~150-word compressed version). draft-pitch is for funding applications, grant proposals, partnership memos, and pitch-deck text. Output is grounded in repo source via check-grounding.py; no overpromising via check-overpromise.py.

Use when the user says "write me an elevator pitch", "give me an elevator pitch", "30-second pitch", "one-paragraph pitch", "quick pitch for [project]", "pitch this in 30 seconds", "short pitch", "compress this to an elevator pitch", or wants a roughly 150-word / 30-seconds-spoken pitch for a project. Output is grounded in repo source via check-grounding.py and avoids overpromising via check-overpromise.py. Different shape from draft-pitch (long-form one-pager / executive summary / deck-equivalent) — elevator-pitch is the tight oral / cold-intro version.

Use when the user says "explain this document", "interpret this file", "what does this plan commit me to", "what does this ADR say", "interpret this architecture doc", "what does this AI-generated plan actually mean", "decompile this doc", "what's this file saying", "explain this to me in plain English", "what are the implications of this doc", "where could this plan go wrong", or wants a specific repo file (a plan, an ADR, an architecture spec, an AI-generated proposal, a contract-shaped doc, a config file) interpreted in plain English with attention to what it commits to, what it assumes, and where it could go wrong.

Use when the user says "audit my README", "ground my README against the code", "check if my README overpromises", "review my README honestly", "rewrite my README without marketing language", "audit my marketplace listing", "check my plugin description", "ground this description against the code", "is my README honest", "does my README match the code", or wants to (a) audit an existing README/marketplace listing against actual code to find overpromise + unbacked claims, or (b) generate a README/listing from code with grounding controls so it doesn't overpromise. Two modes: audit (check existing) and generate (create new).

Use when the user wants to author agentic loops or goal/completion conditions — "write a loop prompt", "loop engineering", "set up a ralph loop", "run Claude overnight", "write a /goal condition", "goal mode", "goal engineering", "stop condition for my agent", "completion criteria", "definition of done for the agent", "long-horizon task setup", "keep the agent working until", "make Codex run until tests pass", "PLANS.md", "harness for a long run", "my agent stops too early", "my agent never finishes", or wants Claude Code /loop, /goal, Stop hooks, or Codex Goal Mode set up well. Covers loop prompt authoring, goal condition design, the four-file long-horizon scaffold, and anti-reward-hacking hardening. For ordinary single-shot prompts, use prompt-engineering instead.

Use when the user says "narrate this project", "explain this project in plain English", "describe what this repo does", "summarize this project for a non-developer", "write a project narrative", "tell me what this project is", "explain this codebase to a collaborator", "draft an onboarding brief", "explain this project to my parents", "non-dev summary", or wants a plain-language description of a project synthesized from its source artifacts (code, docs, planning files, README, package.json). Works on any repo. Adapts to the named audience. Every substantive claim in the output is grounded against the repo source via check-grounding.py; no overpromising via check-overpromise.py. The output is prose for a human reader, not a planning artifact.

Use when the user says "repo-align", "align the repo", "clean up the repo", "check for drift", "is the repo in good shape", "find contradictions", "what's gotten messy", "doc cleanup", "are my docs consistent", "tidy up the docs", or when startup flagged drift. The deep, opt-in alignment pass: finds documentation drift (duplicate authorities, stale docs, contradictions, scope-creep, loose/misplaced docs, broken read paths) and proposes fixes interactively. Proposes before acting on every judgment call; preserves history with git mv; never silently merges, deletes, or archives.

Use when the user says "set up this repo", "bootstrap this project", "new project setup", "scaffold the docs", "initialize the repo model", "get me started right", or when startup recommended it on a fresh repo. Scaffolds the compact doc model (AGENTS.md, docs/prd.md, docs/plan.md, docs/handoff.md, thin agent shims, and minimal folders) in a new or nearly empty repo. Creates only what's missing; never invents product content; never overwrites user-authored files without confirmation.