lklimek
UserPlugin with opinionated set of Claude Code agents nad skills
Categories
Indexed Skills (33)
ci-dance
This skill should be used when the user says 'ci-dance', 'make the PR green', 'ship this and fix CI', 'push and handle reviews', or wants end-to-end PR pipeline automation.
review-dependency
This skill should be used when the user asks to "review a dependency update", "audit this dependency bump", or assess the security of an upgraded or newly added dependency.
triage-findings
This skill should be used when the user explicitly asks to "triage findings" through an interactive browser-based workflow. Only invoke when explicitly requested.
bug-investigation
This skill should be used when investigating a reported bug, diagnosing a failure, or performing root-cause analysis — especially before concluding "not a bug", tracing a symptom, or explaining unexpected behavior. Preloaded on investigation agents.
coding-best-practices
This skill should be used when developing, modifying, reviewing, or testing code. It defines universal rules for TDD, self-review, quality timing, review format, and security. MANDATORY for every agent performing such work — load at task start and apply continuously.
dependabot-merge
This skill should be used when the user asks to "merge dependabot PRs", "process dependency bumps", "auto-merge bot PRs", or "handle the dependabot backlog". It audits each dependency, comments findings, merges when CI is green, and requests rebases for conflicts or CI failures.
frontend-best-practices
This skill should be used when writing, reviewing, or discussing frontend code — TypeScript, React/Vue/Svelte, CSS, accessibility, and testing.
go-best-practices
This skill should be used when writing, reviewing, or discussing Go code — idioms, error handling, concurrency, and testing patterns.
merge-base
This skill should be used when the user asks to "merge the base branch", "update this feature branch from base", or resolve conflicts while merging base into a feature branch.
push
This skill should be used when the user asks to "commit and push", "create a PR", "ship this", "send changes upstream", "open a pull request", or "publish this branch". It commits, pushes, creates a PR, and automatically creates a feature branch when currently on a base branch.
release
This skill should be used when the user asks to "bump the version", "cut a release", or "create a GitHub release". It applies SemVer 2.0, updates the changelog, commits, pushes, creates the release, and auto-detects Rust, Python, JS/TS, Claude Code plugin, and other supported stacks. Arguments are major, minor, patch, or auto-detection from commits. User-invocable only — agents must not invoke it autonomously.
review-loop
Use to autonomously handle peer review — request, wait, fix comments, push, repeat until resolved.
rust-best-practices
This skill should be used when implementing, fixing, writing, reviewing, or discussing Rust code quality, API design, safety, or idioms. It applies the Microsoft Pragmatic Rust Guidelines and Rust API Guidelines checklists.
security-best-practices
This skill should be used when writing or reviewing code that handles authentication, cryptography, user input, secrets, or API endpoints. It provides OWASP-based secure programming practices and should be consulted proactively during reviews and planning.
review-pr
This skill should be used when the user asks to "review this PR", "audit this pull request", or assess a PR for code quality, security, and correctness.
grumpy-review
This skill should be used when the user requests a code review, audit, or quality assessment covering quality, security, dependencies, and documentation. It uses parallel agents and produces a deduplicated, severity-ranked report.
check-pr-comments
This skill should be used when the user asks to "check PR comments", "verify review comments are addressed", or otherwise confirm that PR feedback is resolved in code. It can optionally produce a triage-compatible report.
report-format
This skill should be used when emitting or consuming review findings. It defines the unified review report format for all finding-producing agents.
validate-findings
This skill should be used when a coordinator performs the LLM validation pass on a consolidated v4 findings report. It adds ai_assessment, ai_verdict, and ai_verdict_confidence and, in the rare partial-producer case, re-estimates missing likelihood, impact, and relevance. Coordinator-only.
python-best-practices
This skill should be used when writing, reviewing, or discussing Python code — PEP 8, type hints, testing, error handling, and code quality tooling.
triage
This skill should be used when the user says "triage" or asks to triage a GitHub issue: reproduce it, identify the root cause and owning component, estimate severity, discuss the findings with the user, and then post a short status comment.
git-and-github
This skill should be used when running git or gh commands, interacting with GitHub, or resolving git and gh access or permission-denied failures.
workflow-trivial
Use for typos or single-line fixes (≤20 lines). Same mandatory phase order (Planning→Impl→QA→LL), minimal ceremony. Auto-retry on failure.
spawn-checklist
This skill should be used immediately before any Agent() spawn — a single agent or a whole wave — and whenever the user says "spawn an agent", "delegate this", "parallelize this", "split this work", or "use a subagent". Also applies when choosing a model tier, batching small tasks, or deciding inline-vs-spawn. Reload before every spawn, not once per session.
workflow-feature
This skill should be used when the user asks to "build a new project", "add a feature", or perform a major refactoring. It runs Planning (Req→UX→Test Spec→Dev Plan) → Implementation → QA → Lessons Learned, with unattended auto-retry on failure.
delegate
This skill should be used when preparing to delegate any task to an agent — a single Agent() spawn or a whole wave — or when the user says "spawn an agent", "delegate this", "parallelize this", "split this work", or "use a subagent". It also applies when choosing a model tier, batching small tasks, or deciding inline-vs-spawn. Reload before every delegation, not once per session.
severity
This skill should be used when rating findings in reviews, audits, and assessments. Preloaded on finding-producing agents.
codex-crew
This skill should be used when preparing to dispatch work to Codex Astra, deciding whether to route coding to Codex, dispatching directly via codex-companion.mjs rather than the codex:codex-rescue subagent, handling a Codex job that fails to write or commit, monitoring a running Codex job, or recovering a stale Codex broker. The coordinator reads its pre-flight guidance once before the first Codex dispatch of a session.
grand-admiral
This skill should be used when coordinating multiple agents — spawning subagents, isolating worktrees, managing teams, scaling work, recovering failures, or managing a programme. Coordinator agents that spawn, manage, or merge subagent work always load it.
bye
This skill should be used when the user says "bye", "wrap up", "tear down", "we're done", or otherwise ends the session. It shuts down spawned teammates and agents, removes their worktrees, and reconciles uncommitted or unpushed session work.
lessons-learned
This skill should be used when extracting learnings, saving lessons, or capturing reusable knowledge from a session. It is also appropriate before presenting a plan, after notable events such as bugs or corrected approaches, and as the final task when work is complete.
track-minions
This skill should be used when tracking delegated or multi-step work in a durable store that survives context loss — before spawning agents, while a multi-task wave is in flight, and after compaction or other context loss to recover pending work. It applies to solo, delegated, and multi-agent work alike.
workflow-simplified
This skill should be used when handling a bug fix or small-to-medium change of no more than 1000 LOC. One high-capability agent runs plan → TDD tests → implementation → self-review/self-fix end-to-end from a WHAT-not-HOW brief, unattended.
Bio shown is the top-scored skill's repo description as a fallback — real GitHub bios land in a future update.