genius-skills

Ultra-compressed communication mode. Cuts token usage ~75% by dropping filler, articles, and pleasantries while keeping full technical accuracy. Use when user says "caveman mode", "talk like caveman", "use caveman", "less tokens", "be brief", or invokes /caveman.

Use when a repository, project, service, or workstream needs a name and a descriptive one would be dull — 'what should I call this repo', 'give me a codename', 'name this project', naming a spike, experiment, or internal tool, or replacing a placeholder name. Tuned for repo and internal names, not public brand or trademark work.

Use when the user wants a decision, idea, or piece of work pressure-tested from multiple independent angles before they commit to it — not for questions with one right answer. Based on Karpathy's LLM Council methodology. MANDATORY TRIGGERS: 'council this', 'run the council', 'war room this', 'pressure-test this', 'stress-test this', 'debate this'. STRONG TRIGGERS (when combined with a real decision or tradeoff): 'should I X or Y', 'which option', 'what would you do', 'is this the right move', 'validate this', 'get multiple perspectives', 'I can't decide', 'I'm torn between'. Do NOT trigger on simple yes/no questions, factual lookups, or a casual 'should I' without a meaningful tradeoff (e.g. 'should I use markdown'). DO trigger when the user presents a genuine decision with stakes, multiple options, and context suggesting they want it stress-tested from multiple angles.

Disciplined diagnosis loop for hard bugs and performance regressions. Reproduce → minimise → hypothesise → instrument → fix → regression-test. Use when user says "diagnose this" / "debug this", reports a bug, says something is broken/throwing/failing, or describes a performance regression.

Interview the user relentlessly about a plan or design until reaching shared understanding, resolving each branch of the decision tree one at a time with a recommended answer for each. When the project carries domain docs (CONTEXT.md / ADRs), also challenge the plan against them, sharpen terminology, and update those docs inline. Use to stress-test a plan or design, or when the user says "grill me".

Compact the current conversation into a handoff document for another agent to pick up.

Find deepening opportunities in a codebase, informed by the domain language in CONTEXT.md and the decisions in docs/adr/. Use when the user wants to improve architecture, find refactoring opportunities, consolidate tightly-coupled modules, or make a codebase more testable and AI-navigable.

Build a throwaway prototype to flush out a design before committing to it. Routes between two branches — a runnable terminal app for state/business-logic questions, or several radically different UI variations toggleable from one route. Use when the user wants to prototype, sanity-check a data model or state machine, mock up a UI, explore design options, or says "prototype this", "let me play with it", "try a few designs".

Use when you need a recording or visual proof of a real user journey through a running app — a demo video for a PR, ticket or stakeholder, evidence that a fix behaves correctly end to end, a walkthrough captured at a fixed device viewport (TV, tablet, kiosk), or the same journey re-recorded later after code changed. Also use when a recording came out cropped, juddering, or showing something that did not happen.

Test-driven development with red-green-refactor loop. Use when user wants to build features or fix bugs using TDD, mentions "red-green-refactor", wants integration tests, or asks for test-first development.

Triage issues through a state machine driven by triage roles. Use when user wants to create an issue, triage issues, review incoming bugs or feature requests, prepare issues for an AFK agent, or manage issue workflow.