Git 与版本控制
Git 操作、提交助手与分支管理
specToProvenPR
jthorare
Use when turning an approved spec or design doc into production-ready pull requests that are definitively proven to work in the running app. Use for staged multi-phase builds where each stage must be planned, implemented, tested by hand in the running app, evidenced, and only then driven to zero blocking review findings. Triggers include "take this spec to PRs", "ship this design", "prove it works then open the PR", staged epic delivery, and review-loop-until-clean. Also use when tempted to stop a review loop early, treat green tests as proof, or defer findings to a follow-up.
seat-workspace-identity
musterd-io
Give every agent its own workspace with its own git identity, so "who wrote this commit" is answerable without asking anyone — and know the exact line where that goes wrong, because repo-local git config is shared across work trees and the last one written owns everybody's commits. Ships a checker that tells a scoped identity from a shared one. Use when running more than one agent on a repo, when commits carry the wrong author, or when setting up agent workspaces.
conventional-commits
bastos
Format commit messages using the Conventional Commits specification. Use when creating commits, writing commit messages, or when the user mentions commits, git commits, or commit messages. Ensures commits follow the standard format for automated tooling, changelog generation, and semantic versioning.
thumbgate-guard
IgorGanapolsky
Turn the agent's most recent mistake into an enforced ThumbGate prevention rule (a PreToolUse block gate) so the same bad tool call is intercepted before it runs again, in this and every future session across Claude Code, Cursor, Codex, Gemini, Amp, and Cline. Captures the failure with the capture_feedback MCP tool, then force-promotes it via npx thumbgate force-gate so it is enforced, not just logged. Use when the user says "guard against this", "block this from happening again", "never do that again", "make that a rule", "stop the agent from repeating that", or right after a bad action or thumbs-down that should become a hard rule. Do NOT use to merely log a thumbs-up/down without enforcement (use the thumbgate-feedback skill), to recall prior context before starting work (use the Agent Memory skill), or to list rules that already exist (use the thumbgate-rules skill).
ci-failure-triage
IgorGanapolsky
Triage a CI / PR check failure by READING the failure body before forming any hypothesis. Triggered whenever a required check is red, a PR is BLOCKED, a merge won't land, or you're about to call a failure "transient", "flaky", "stale", or "orphaned". Prevents dismissing a real failure (e.g. real CodeQL security alerts) as noise.
shipRelease
jthorare
"Ship a release — sync branch, run tests, audit coverage, push, open PR, then auto-chain → /landAndDeploy → /canary → /syncDocs (skip the deploy chain with --no-deploy)."
get-latest-ci
nrwl
Fetch the latest CI pipeline execution for the current branch. Returns the most recent CIPE which may be completed, in progress, or null. Use when you need to review CI status, check failures, or inspect CI state.
dreaming
musterd-io
How a seat consolidates overnight — save the day's traps and numbers as insights, promote anything seen twice into a wiki page with a date and a falsifier, as a PR that wiki:check gates. Use when Presence provenance is dream, or when you hold the dreaming write path.
decision-records
musterd-io
The whole lifecycle of a numbered decision record — taking a number against everything in flight rather than everything that landed, publishing it before you write, freezing the Decision on accept so amendments are dated and append-only, and naming what would tell you it was wrong. Ships a checker that refuses a rewritten Decision, including across a rename. Use when adopting ADRs, when two records collided on a number, or when someone edits a decision that already shipped.
autonomous-execution-mandate
IgorGanapolsky
Self-directed autonomous execution mandate and workflow skill for ThumbGate agents. Enforces zero-human-intervention task chaining, proactive CI/Trunk monitoring, automated conflict resolution, and evidence-first reporting without requiring manual prompt nudges from the user.
thumbgate-protect
IgorGanapolsky
Inspect this repo's branch and release governance (protected branches, release rules, protected-file globs) and, only when the user explicitly approves, grant a scoped, time-limited exception so a protected-file edit or publish can proceed under audit. Reads posture via the get_branch_governance MCP tool and records a narrow, expiring approval via the approve_protected_action MCP tool. Use when the user says "is main protected", "show branch governance", "what am I blocked from editing", "approve this protected change", or "let me edit a protected file just this once". Do NOT use to disable protection wholesale, to grant broad or standing exceptions, or to diagnose hook wiring (use the thumbgate-doctor skill) — this skill is for narrow, temporary, audited approvals only.
Skill: ApplyOps Deploy
IgorGanapolsky
Plausible domain is igorganapolsky.github.io, NOT thumbgate-production.up.railway.app.
studio-deliver
azerothl
Préparer et livrer une modification Git demandée avec branche, commit, PR et état CI traçables, selon les actions externes autorisées.
measure-agents-honestly
musterd-io
How to measure whether a multi-agent setup actually helps — a pre-registered manifest whose ruler cannot bend to fit the result, wasted work reconstructed from git alone, results that diagnose rather than rank, and the denominator that decides whether your headline is a finding or a sales pitch. Use before running any agent comparison, when writing up one, or when someone quotes a multiplier at you.
commit-changes
SamGu-NRX
Clean up, review, validate, and commit a repository's intended changes. Use when asked to commit, make a commit, write a commit message and commit, or finish local changes as a Git commit. Always invoke $deslop across every staged file before drafting the message or running git commit.
babysit
SamGu-NRX
'Watch a PR or related PR stack until merge or closure, coordinate CI and review fixes, and recommend human review. Use for "babysit #N", "watch this stack", or "keep these PRs green". Never merges.'
info
osdlc
Report the runtime state of the sdlc plugin this session is actually running — the on-disk root it loaded from (CLAUDE_PLUGIN_ROOT, resolved through any symlink), the manifest version, whether it's a live working tree or a pinned cache snapshot, the git branch/commit/dirty state, and every ~/.claude/plugins registration that mentions the plugin. Read-only; for self-verifying which install won when several are configured.
interminai
mstsirkin
Control interactive terminal applications like vim, git rebase -i, git add -i, git add -p, apt, rclone config, sudo, w3m, and TUI apps. Can also supervise another CLI LLM (cursor-agent, codex, etc.) - approve or reject its actions by pressing y/n at confirmation prompts. Use when you need to interact with applications that require keyboard input, show prompts, menus, or have full-screen interfaces. Also use when commands fail or hang with errors like "Input is not a terminal" or "Output is not a terminal". Better than application specific hacks such as GIT_SEQUENCE_EDITOR or bypassing interactivity through file use.
brainstorming
CommandCodeAI
"You MUST use this before any creative work - creating features, building components, adding functionality, or modifying behavior. Explores user intent, requirements and design before implementation."
delegate-to-zcode
zrll2
Delegate code writing to ZCode (GLM) while Claude Code decides, verifies and reviews. Use when the user says "delegate to zcode", "let zcode write it", "zcode do this", asks for parallel or autonomous coding, or when a project's CLAUDE.md says to delegate code writing to ZCode. Covers the full loop: preflight, git worktree, work order, run, verify, fix order, review, commit.
code-review
unix2dos
codereview / code review / 代码审查 - Go 代码审查技能，专注于性能、并发安全、安全性和可读性四大核心维度。
learn-tech
unix2dos
技术知识学习助手 - 帮助用户以结构化方式深入学习技术概念。扮演耐心老师角色，用通俗易懂的方式讲解复杂概念。触发方式：@learn-tech {主题} 或 "学习 {主题}"、"讲解 {主题}"、"teach me {topic}" 等。输出包含核心概念、实践要点、知识拓展、互动检查四大模块，强制生成 Mermaid 图表，自动保存到 learn-tech_outputs 目录。
specstory-guard
specstoryai
Install a pre-commit hook that scans .specstory/history for secrets before commits. Run when user says "set up secret scanning", "install specstory guard", "protect my history", or "check for secrets".
pm-git-file-tracking
bobmatnyc
Protocol for tracking files immediately after agent creation