Agent
Part of kyzo-plan
RUN FIRST after startstory — delete Condemned paths only. Red tree OK; preservation fails. Parent arms condemned allowlist and path-monitors. No Bash, git, or build. Not development or judge.
2,163 tagged Git and workflow, measured the same way as everything else here.
Browse within: ai-coding 97agent-orchestration 66Multi-Agent 39code-quality 37Autonomous Agents 35claude-ai 35agentic-workflow 32copilot 31ai-assistant 29agentic-coding 27ai-workflow 27antigravity 25ai-governance 24AI Safety 23
Agent
Part of kyzo-plan
RUN FIRST after startstory — delete Condemned paths only. Red tree OK; preservation fails. Parent arms condemned allowlist and path-monitors. No Bash, git, or build. Not development or judge.
Agent
Part of kyzo-plan
Execute ONE story T# under allowlist — Edit/Write only, then kyzo-plan-task-completion-request to the judge. After demolition. Spawn via taskspawn XML. No Bash, git, or board mutate. Not demolition or judge.
Agent
Generate changelog entries from git history between versions.
Agent
Bump version in package.json, src/index.mjs, CHANGELOG.md, and README.md.
crcresearch/llm-wiki-memory-template
Agent
Canonical pre-commit check for any wiki write. The Verification Gate is a small, project-extensible criteria list that every ingest skill (/wiki-experiment, /wiki-source, or any future write skill) runs before staging and committing.
crcresearch/llm-wiki-memory-template
Agent
Canonical procedure for pushing changes to the wiki sub-repo when multiple agents may be writing concurrently. The reference implementation at scripts/wiki-write-protocol/protocol.sh handles the optimistic-push-with-retry mechanics (CI-validated on 9 scenarios); this file is the agent-facing procedure that tells…
Agent Claude Code
Use as the FINAL stage of a proposal-to-pr run to record what happened, as one retrospective file under .claude/retrospectives/. Dispatched fresh, having NOT done the work, so it reads artefacts — the pipeline state file, review-evidence.md and its ## Probe — stage 2 score, progress.md, git history — rather than the…
open-horizon-labs/repo-native-alignment
Agent Claude Code
RNA delivery pipeline from implementation to merge, with delivery verification, final comment sweep, and independent final-diff approval.
Emasoft/claude-plugins-validation
Agent
Part of claude-plugins-validation
Creates new Claude Code plugin or marketplace repositories from scratch with full CI/CD pipeline, git hooks, and standard file structure, then publishes and watches GitHub CI to green before declaring done. Use when user wants to create a new plugin, scaffold a repo, or set up a new marketplace hub.
Agent
Part of dev-workflow
Reviews code changes on the current task branch and returns severity-ranked findings under an orchestrator.
andronics/claude-plugin-typescript-pro
Agent
Part of typescript-pro
TypeScript library packaging and npm publishing specialist. Use for preparing libraries for npm, generating declaration files, configuring dual packages (ESM/CJS), and managing library releases.
Agent
Part of fakoli-plugin-critic
Use this agent when you need a cross-plugin structural review — plugin.json manifest, marketplace.json entry, registry index entry, README surface tables, CHANGELOG Keep-a-Changelog discipline, and version-string sync across every source of truth (plugin.json, pyproject.toml, init.py, marketplace, registry). Adapts…
CurtisThe/three-pillars-plugin
Agent
Part of three-pillars
Write-capable worker subagent for the autonomous orchestrator. Implements a plan on a candidate branch, commits, pushes, and returns the candidate.v1 envelope.
Agent
Part of recce
Single-environment Tier-1 verification specialist for dbt model changes. Dispatched by /recce-verify when the user has no target-base/ artifacts and wants a lightweight pre-commit risk read. Combines column-level lineage (getcll), AST/semantic analysis (analyzemodel), and targeted current-env SQL probes (query) to…
Agent
Part of df
Independent, isolated code reviewer. Use to review a finished work product (a git diff against a plan/spec) for spec compliance and code quality before committing. Receives ONLY a path to the diff file, a path to the spec/plan and its acceptance criteria — never the implementation conversation. Returns a…
Agent
Part of ghe
Use this agent when validating phase transitions in GitHub Elements workflow. Ensures DEV to TEST to REVIEW order is maintained, checks prerequisites before transitions, and blocks invalid transitions. Use before any phase transition, when validating workflow state, or checking if transition is allowed. Examples…
Agent
Use PROACTIVELY when user mentions: review, code review, audit, quality, assess, evaluate, critique, feedback, score, rate, check my code, how does this look, is this good, technical debt, code smell, best practices, clean code, code quality, maintainability, refactor review, PR review, pull request, merge request…
Agent
Part of cadence
Babysit a pull request until it settles — green required CI and no unresolved actionable review threads; triage, fix, push, repeat.
Agent
An agent for managing Git work in the MCP Tool Hub project. Git is a system for tracking code changes, branches, versions, and releases.
Agent Claude Code
CI/CD & release-engineering currency expert. Researches latest GitHub Actions, runner images, CodeQL, base-image, and MCP-registry publishing practices on official sources and compares them against this repo's workflows and Dockerfile. Use for the ci slice of a currency review, or ad-hoc "are our workflows/actions…
Agent Claude Code
Part of ivuorinen-skills
Verifies a release is ready to ship — checks skill and command validity, findings store, version sync, conventional commits, and CI status.
Agent
Review a change against the history of the code it touches. Use git blame and log to catch changes that contradict why the surrounding code exists — reintroducing a fixed bug, removing a guard that was added on purpose, or undoing a deliberate decision. Reports only high-confidence findings backed by a specific past…
Agent
Review a change against prior pull requests. Use the PR/commit history to catch an approach that was already tried, discussed, or reverted — so the change does not re-litigate a settled decision or redo work that was rejected. Reports only high-confidence findings backed by a specific prior PR or revert.
Agent
Part of sw
The sole owner of one specwright change dispatched by a delivery: runs the change ladder end to end in its own branch and worktree, records the decisions it took, and reports shipped or blocked back to the delivery orchestrator.
At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: