tobihagemann/turbo

这个仓库里有 153 个技能，GitHub 星标 ★ 403。

• answer-reviewer-questionsFor each reviewer question on a PR, recall implementation reasoning and compose
• apply-findingsApply findings by making the suggested code changes. Applies accepted verdicts,
• assess-technical-debtAssess project-wide structural technical debt: complexity hotspots, deprecated A
• auditProject-wide health audit pipeline that fans out to all analysis skills in paral
• changelog-rulesShared changelog conventions and formatting rules referenced by /create-changelo
• code-styleEnforce existence, reuse, mirror, and symmetry principles to keep new code minim
• codex-execRun autonomous task execution using the codex CLI. Use when the user asks to \"c
• codex-reviewRun AI-powered code review using the codex CLI. Use when the user asks to \"code
• commit-rulesShared commit message rules and technical constraints referenced by /stage-commi
• commit-stagedCommit already-staged changes with a message matching existing commit style. Use
• commit-staged-pushCommit already-staged changes and push in one step. Use when the user asks to \"
• consult-codexMulti-turn consultation with Codex CLI for second opinions, brainstorming, or co
• consult-oracleConsult ChatGPT Pro via ChatGPT browser automation for problems that resist stan
• contribute-turboPropose a turbo skill improvement upstream by filing a GitHub issue against the
• create-changelogCreate a CHANGELOG.md following keepachangelog.com conventions with version hist
• create-handoffWrite a handoff file at .turbo/handoff/<YYYY-MM-DD>-<slug>.md capturing current
• create-issueCreate a GitHub issue with a drafted title and body. Use when the user asks to \
• create-prCreate a GitHub pull request with a drafted title and description. Use when the
• create-project-skillsScans an existing codebase and generates project-specific skills that capture in
• create-skillCreate a new skill or update an existing skill that extends Claude's capabilitie
• create-test-planAnalyze what changed and generate a structured test plan at .turbo/test-plans/<s
• create-threat-modelAnalyze a codebase and produce a structured threat model at .turbo/threat-model.
• discuss-changeAlign on the shape of a change through an interview, then implement it. Escalate
• draft-planProduce an implementation plan at .turbo/plans/<slug>.md. Use when the user asks
• evaluate-findingsCritically assess external feedback (code reviews, AI reviewers, PR comments) an
• explain-thisExplain whatever the user is pointing at right now in plain language: a pending
• exploratory-testExecute multi-level exploratory testing of the app covering basic functionality,
• fetch-pr-commentsFetch and summarize review feedback and conversation from a GitHub PR (unresolve
• finalizeRun the post-implementation quality assurance workflow including tests, code pol
• find-dead-codeFind dead code using parallel subagent analysis and optional CLI tools, treating
• frontend-designCreate distinctive, production-grade frontend interfaces with high design qualit
• github-voiceShared writing style rules for GitHub-facing output (PR comments, PR description
• implementLoad code-style and task-specific skills, make the change described by the curre
• implement-improvementsValidate improvements from .turbo/improvements.md, recommend a working set tailo
• implement-planExecute an implementation plan file produced by /draft-plan or /turboplan. Runs
• interpret-feedbackInterpret third-party feedback by running parallel internal and peer interpretat
• investigateSystematically investigate bugs, test failures, build errors, performance issues
• map-codebaseDeep architecture report that fans out parallel inspections across different asp
• note-improvementCapture an out-of-scope improvement opportunity so it doesn't get lost. Use when
• onboardDeveloper onboarding guide that composes architecture mapping, tooling review, a
• peer-reviewRun an independent peer review via Codex. Use when the user asks to \"peer revie
• pick-next-issueFetch and rank open GitHub issues by community engagement, present the top 3 can
• polish-codeStage, format, lint, test, review, smoke test, and re-run itself until stable. U
• previewStand up the project's live app and hand it to the user to try a change firsthan
• prototypeBuild a self-contained local prototype at .turbo/prototypes/<slug>.html, drive i
• quick-finalizeClose out a change without the deep review loop: stage, simplify code and docs,
• recall-reasoningRecall the reasoning behind a past change by locating the Claude Code transcript
• refine-planIteratively review and revise a plan until no new findings survive evaluation. U
• reply-to-pr-conversationDraft, confirm, and post a single conversational reply to GitHub PR conversation
• reply-to-pr-threadsDraft, confirm, and post replies to GitHub PR review threads. Handles per-catego
• resolve-findingsChoose an implementation path (direct or plan) for evaluated findings and dispat
• resolve-pr-commentsEvaluate, fix, answer, and reply to GitHub pull request review comments and conv
• review-agentic-setupDetect agentic coding infrastructure in a project: CLAUDE.md, AGENTS.md, install
• review-codeReview code for bugs, security vulnerabilities, API misuse, consistency issues,
• review-dependenciesDetect package managers and CI action pins, then discover outdated or vulnerable
• review-planReview a plan by running internal reviews and a peer review in parallel and retu
• review-prReview a pull request by fetching PR comments, running a comprehensive code revi
• review-toolingDetect what dev tooling infrastructure a project has and flag gaps across linter
• run-checksRun the project's full verification gate: every check the project defines as a p
• self-improveExtract lessons from the current session, or sweep the project's past sessions w
• shipCommit, push, and optionally create or update a PR for the current staged change
• simplify-allRun the code and documentation simplification passes together over one scope, la
• simplify-codeRun a multi-agent review of changed files for scope, reuse, quality, efficiency,
• simplify-docsRun a multi-agent review of code comments and markdown documentation for unneces
• smoke-testLaunch the app and hands-on verify that it works by interacting with it. Falls b
• split-and-shipExecute an approved split plan, shipping each change group separately as its own
• stageStage implementation changes for commit with precise file selection. Use when th
• stage-commitStage files and create a commit in one step with a message matching existing com
• stage-commit-pushStage files, create a commit, and push in one step. Use when the user asks to \"
• survey-patternsSurvey the codebase for analogous features, reusable utilities, and existing pat
• turboplanAnalyze task complexity and route to a mode by artifact: direct fix for clear-sc
• understand-changeTeach the user to deeply understand a change through interactive tutoring: resta
• update-changelogUpdate the Unreleased section of CHANGELOG.md based on current changes. No-op if
• update-dependenciesUpgrade project dependencies with breaking change research for major version upd
• update-prUpdate an existing GitHub pull request's title and description to reflect the cu
• update-turboUpdate installed Turbo skills from the local repo with a dynamic changelog, conf
• user-experienceApply a UX lens to a user-facing change: whether it serves the user's real goal
• answer-reviewer-questionsFor each reviewer question on a PR, recall implementation reasoning and compose
• apply-findingsApply findings by making the suggested code changes. Applies accepted verdicts,
• assess-technical-debtAssess project-wide structural technical debt: complexity hotspots, deprecated A
• auditProject-wide health audit pipeline that fans out to all analysis skills in paral
• changelog-rulesShared changelog conventions and formatting rules referenced by $create-changelo
• claude-printRun a non-interactive Claude Code print-mode call from Codex. Use when the user
• code-styleEnforce existence, reuse, mirror, and symmetry principles to keep new code minim
• commit-rulesShared commit message rules and technical constraints referenced by $stage-commi
• commit-stagedCommit already-staged changes with a message matching existing commit style. Use
• commit-staged-pushCommit already-staged changes and push in one step. Use when the user asks to \"
• consult-claudeConsult Claude Code for second opinions, brainstorming, or difficult debugging f
• consult-oracleConsult ChatGPT Pro via ChatGPT browser automation for problems that resist stan
• contribute-turboPropose a turbo skill improvement upstream by filing a GitHub issue against the
• create-changelogCreate a CHANGELOG.md following keepachangelog.com conventions with version hist
• create-handoffWrite a handoff file at .turbo/handoff/<YYYY-MM-DD>-<slug>.md capturing current
• create-issueCreate a GitHub issue with a drafted title and body. Use when the user asks to \
• create-prCreate a GitHub pull request with a drafted title and description. Use when the
• create-project-skillsScans an existing codebase and generates project-specific skills that capture in
• create-skillCreate a new skill or update an existing skill that extends Codex's capabilities
• create-test-planAnalyze what changed and generate a structured test plan at .turbo/test-plans/<s
• create-threat-modelAnalyze a codebase and produce a structured threat model at .turbo/threat-model.
• discuss-changeAlign on the shape of a change through an interview, then implement it. Escalate
• draft-planProduce an implementation plan at .turbo/plans/<slug>.md. Use when the user asks
• evaluate-findingsCritically assess external feedback (code reviews, AI reviewers, PR comments) an
• explain-thisExplain whatever the user is pointing at right now in plain language: a pending
• exploratory-testExecute multi-level exploratory testing of the app covering basic functionality,
• fetch-pr-commentsFetch and summarize review feedback and conversation from a GitHub PR (unresolve
• finalizeRun the post-implementation quality assurance workflow including tests, code pol
• find-dead-codeFind dead code using parallel sub-agent analysis and optional CLI tools, treatin
• frontend-designCreate distinctive, production-grade frontend interfaces with high design qualit
• github-voiceShared writing style rules for GitHub-facing output (PR comments, PR description
• implementLoad code-style and task-specific skills, make the change described by the curre
• implement-improvementsValidate improvements from .turbo/improvements.md, recommend a working set tailo
• implement-planExecute an implementation plan file produced by $draft-plan or $turboplan. Runs
• interpret-feedbackInterpret third-party feedback by running parallel internal and peer interpretat
• investigateSystematically investigate bugs, test failures, build errors, performance issues
• map-codebaseDeep architecture report that fans out parallel inspections across different asp
• note-improvementCapture an out-of-scope improvement opportunity so it doesn't get lost. Use when
• onboardDeveloper onboarding guide that composes architecture mapping, tooling review, a
• peer-reviewRun an independent peer review via Claude. Use when the user asks to \"peer revi
• pick-next-issueFetch and rank open GitHub issues by community engagement, present the top 3 can
• polish-codeStage, format, lint, test, review, smoke test, and re-run itself until stable. U
• previewStand up the project's live app and hand it to the user to try a change firsthan
• prototypeBuild a self-contained local prototype at .turbo/prototypes/<slug>.html, drive i
• quick-finalizeClose out a change without the deep review loop: stage, simplify code and docs,
• recall-reasoningRecall the reasoning behind a past change from Codex session history when availa
• refine-planIteratively review and revise a plan until no new findings survive evaluation. U
• reply-to-pr-conversationDraft, confirm, and post a single conversational reply to GitHub PR conversation
• reply-to-pr-threadsDraft, confirm, and post replies to GitHub PR review threads. Handles per-catego
• resolve-findingsChoose an implementation path (direct or plan) for evaluated findings and dispat
• resolve-pr-commentsEvaluate, fix, answer, and reply to GitHub pull request review comments and conv
• review-agentic-setupDetect agentic coding infrastructure in a project: CLAUDE.md, AGENTS.md, install
• review-codeReview code for bugs, security vulnerabilities, API misuse, consistency issues,
• review-dependenciesDetect package managers and CI action pins, then discover outdated or vulnerable
• review-planReview a plan by running internal reviews and a peer review in parallel and retu
• review-prReview a pull request by fetching PR comments, running a comprehensive code revi
• review-toolingDetect what dev tooling infrastructure a project has and flag gaps across linter
• run-checksRun the project's full verification gate: every check the project defines as a p
• self-improveExtract lessons from the current session, or sweep the project's past sessions w
• shipCommit, push, and optionally create or update a PR for the current staged change
• simplify-allRun the code and documentation simplification passes together over one scope, la
• simplify-codeRun a multi-agent review of changed files for scope, reuse, quality, efficiency,
• simplify-docsRun a multi-agent review of code comments and markdown documentation for unneces
• smoke-testLaunch the app and hands-on verify that it works by interacting with it. Falls b
• split-and-shipExecute an approved split plan, shipping each change group separately as its own
• stageStage implementation changes for commit with precise file selection. Use when th
• stage-commitStage files and create a commit in one step with a message matching existing com
• stage-commit-pushStage files, create a commit, and push in one step. Use when the user asks to \"
• survey-patternsSurvey the codebase for analogous features, reusable utilities, and existing pat
• turboplanAnalyze task complexity and route to a mode by artifact: direct fix for clear-sc
• understand-changeTeach the user to deeply understand a change through interactive tutoring: resta
• update-changelogUpdate the Unreleased section of CHANGELOG.md based on current changes. No-op if
• update-dependenciesUpgrade project dependencies with breaking change research for major version upd
• update-prUpdate an existing GitHub pull request's title and description to reflect the cu
• update-turboUpdate installed Turbo Codex skills from the local repo with a dynamic changelog
• user-experienceApply a UX lens to a user-facing change: whether it serves the user's real goal