Loading...
Loading...
Found 21 Skills
Loads orchestrate mode — a disciplined delivery loop that enforces BDD specs in specs/, real integration tests (no mocks), PR CI and CodeRabbit babysitting, and mandatory end-user QA via computer-use or CLI dogfooding before anything is considered done. Use when starting any non-trivial implementation task, feature build, or delivery where you want the work driven from spec to proven-shipped state rather than stopping at "tests pass".
Drive a PR to mergeable state — check CodeRabbit and reviewer comments, address valid ones, reply and resolve invalid ones, fix CI failures (even pre-existing ones from main), push fixes, then loop every 4.5 minutes until two consecutive clean checks. Use after opening a PR when you want it babysitted to green.
Use when reviewing a PR with contrarian inversion to stress-test changes via @coderabbitai, making specific factual claims about dependency behavior that Conejo can later verify by reading library source code. Triggers on proud-zanahoria, contrarian review, inverse review, devil's advocate PR, zanahoria.
Run comprehensive PR checks including reviewing CodeRabbit comments, ensuring PR description quality, running pre-commit hooks, tests, and validation. Use on an existing PR to address review feedback.
Use when the user wants a plan validated by @coderabbitai but the right approach is uncertain — file 2-3 parallel issues with the SAME GOAL but deliberately shuffled assumptions/layers/triggers, so CR has comparative material instead of yes/no on a single approach. Triggers on multi-assumptions, multi-framing, parallel plans, comparative plan, dangle-the-carrots.
Use after zanahoria-multi-assumptions has filed N parallel issue variants AND @coderabbitai has responded to all of them — closes the family by extracting the load-bearing assumption, naming a winner, capturing the decision as an ADR, and cleanly closing the rejected variants with cross-referenced reasoning. Triggers on close-the-family, pick-a-winner, commit-the-decision, decide-variants, after-CR-responds, zanahoria-decisions.
Fetch ALL unresolved PR conversations — review threads, review body comments, and regular PR comments — then output a strict /tdd plan with atomic commits that fixes and replies to every one. Use this skill whenever the user wants to resolve, address, or fix PR review feedback, comments, threads, nitpicks, or suggestions from any reviewer (human or bot like CodeRabbit, Copilot, etc.). Triggers on phrases like "resolve PR threads", "fix PR comments", "address review feedback", "resolve conversations".
Respond to PR review comments with critical evaluation. Use when addressing code review feedback, responding to bot review comments (Gemini Code Assist, CodeRabbit, etc.), or handling PR suggestions. Fetches comments, evaluates each against project context, applies valid fixes, declines invalid suggestions with reasoning, and posts responses.
Use this skill to process bot review feedback on a pull request (for example coderabbitai and chatgpt-codex-operator). Invoke when bot comments arrive and the user wants triage, decision making, and implementation of required fixes.