All authors
Claude Skills by tony-adamson
github.com/tony-adamson14 skills0 installs6 views
- Codebase AnalysisUse when explicitly asked for $codebase-analysis to create or fully update CURRENT_STATE.md with evidence for the current repository or focused area. Do not use for casual code questions, reviews, implementation, debugging, solution design, or planning.Votes: 0GitHub stars: 3
- DebugUse when explicitly asked for $debug, or when a bug survived the first fix, is intermittent, or is a performance regression: build a feedback loop that goes red on the bug, minimise, rank falsifiable hypotheses, instrument, fix with a regression test at a correct seam, clean up. Do not use for a first-read error that the stack trace already explains.Votes: 0GitHub stars: 3
- Ops ReviewUse when explicitly asked for $ops-review to review an implemented change for silent operational failures: missing timeouts, unbounded resources, connection leaks, non-idempotent retries, slow-dependency degradation. Read-only findings report; does not fix code and does not replace load testing.Votes: 0GitHub stars: 3
- Planf3Use when explicitly asked for $planf3 to create, update, or build a minimal Markdown implementation plan from SOLUTION.md or a concrete approved task. Do not use for solution design or speculative architecture.Votes: 0GitHub stars: 3
- Scope ReviewUse when explicitly asked for $scope-review to review a diff for excess scope before a PR: additions without a forcing requirement, foreign patterns, indirection layers, duplication, files outside the task frame. Read-only findings report; does not fix code.Votes: 0GitHub stars: 3
- Solution DesignUse when explicitly asked for $solution-design to create SOLUTION.md for one concrete software change before implementation planning. Produces a minimal evidence-based design; does not write code or create an implementation plan.Votes: 0GitHub stars: 3
- Codebase AnalysisCreates CURRENT_STATE.md: an evidence-backed description of the current state of the repository. Use only when the user explicitly asks to run codebase-analysis. Does not design a solution, write code, or produce an implementation plan.Votes: 0GitHub stars: 3
- DebugDisciplined diagnosis loop for hard bugs, flaky failures and performance regressions: build a feedback loop that goes red on this bug → minimise → rank hypotheses → instrument → fix with a regression test → clean up. Use when the user says debug/diagnose, or reports something broken, throwing, failing intermittently or slow, and the first read of the error did not fix it.Votes: 0GitHub stars: 3
- Ops ReviewReviews an implemented change for silent operational failures: missing timeouts, unbounded resources, connection leaks, non-idempotent retries, slow-dependency degradation. Read-only, findings only, no fixes. Use only when the user explicitly asks to run ops-review.Votes: 0GitHub stars: 3
- Planf3Creates or executes a minimal Markdown implementation plan based on SOLUTION.md or an explicit request. Use only when the user explicitly asks: planning or Build Plan. No presentation artifacts and no future-proof architecture.Votes: 0GitHub stars: 3
- Scope ReviewРевью обоснованности diff перед PR: каждое добавление вызвано текущим требованием, код мимикрирует под локальные паттерны, нет прослоек и дублирования, diff гигиеничен. Только находки, без правок. Запускать только по явной просьбе пользователя.Votes: 0GitHub stars: 3
- Solution DesignCreates SOLUTION.md for one concrete task: the minimal sufficient solution, grounded in the requirements, CURRENT_STATE.md, and live code. Use only when the user explicitly asks to run solution-design. Does not write code and does not create an implementation plan.Votes: 0GitHub stars: 3
- GrillUse when explicitly asked for $grill to interview the user about a task before solution design: works the decision tree in rounds, finds facts itself, puts every decision to the user, records each answer verbatim in GRILL.md. Does not design, plan, or write code.Votes: 0GitHub stars: 3
- GrillInterviews the user until the task is understood in detail, before any design: works the decision tree in rounds, finds facts itself, puts every decision to the user, and records each answer verbatim in GRILL.md. Use only when the user explicitly asks to run grill. Does not design, plan, or write code.Votes: 0GitHub stars: 3