leek-skills

Plugin leek-skills from marketplace leek-agent-skills in leek/agent-skills. Commands read from the repo's manifest — yesterday.

Skills (56)

• architecture-satisfaction

  Run a bounded architectural refactor loop toward a stated design.

• autopilot

  Drive an existing wayfinder map, spec, or set of tickets to completion through fresh top-level Claude Code, Codex, or Grok sessions.

• browser-test

  Test a web app from a start URL and a plain-language prompt, and get one verdict table. Read-only unless you confirm a write.

• chatter-scout

  Sweep the web for new chatter on a topic, skip what earlier runs already saw, verify what is left, and write one ranked, cited findings file to feed a blog post or a digest email.

• closeout

  Check whether this session is safe to close: unsaved work, running jobs, open PRs, unanswered questions, and steps still to run. Each open item names its fix, and closeout runs the fixes the user picks.

• code-review

  Three-axis review of the diff since a fixed point: Correctness (bugs, failure modes, security), Standards (repo conventions) and Spec (ticket/PRD fidelity). Use when the user wants to review a branch, PR, WIP changes, or asks to \"review since X\".

• code-smells-audit

  Audit a codebase, path, or branch diff for classic code smells against the 56-smell Luzkan catalog (PHP/Laravel and TS/React heuristics). Use when the user asks for a code-smell audit or smell-check of a module or diff.

• codebase-design

  Shared vocabulary for designing deep modules in any codebase. Use when designing or improving a class or module's interface, finding deepening opportunities, deciding where a seam goes, making code more testable or AI-navigable, or when another skill needs the deep-module vocabulary.

• commit

  Commit the work done in this session, skipping files a parallel session also touched.

• dependency-audit

  Run a dependency audit for Composer or npm projects and propose a safe upgrade plan.

• diagnosing-bugs

  Diagnosis loop for hard bugs and performance regressions. Use when the user says "diagnose"/"debug this", or reports something broken/throwing/failing/slow.

• domain-modeling

  Build and sharpen a project's domain model: its GLOSSARY.md and ADRs. Use when pinning down domain terminology or a ubiquitous language, challenging fuzzy terms, recording an architectural decision, or when another skill needs the domain model maintained.

• fix-ci-failures

  Read a branch's latest failed CI runs, triage each failure from its existing logs, and fix the real ones.

• fix-github-issues

  Verify, triage, fix, and close review-bot GitHub issues one finding at a time.

• fix-nightwatch-issues

  Triage open Laravel Nightwatch production issues by root cause, fix the legit ones, and resolve them with evidence.

• fix-posthog-issues

  Triage active PostHog error-tracking issues by root cause, fix the legit ones, and resolve or suppress the rest with evidence.

• grill-me

  Run a grilling session: the user wants their plan, decision, or idea stress-tested one question at a time.

• grill-with-docs

  Run a grilling session with domain-modeling alongside, capturing terms in GLOSSARY.md and decisions as ADRs while they land. The default front door for planning work in an existing codebase.

• grilling

  Interview the user one decision at a time until reaching shared understanding, resolving each branch of the decision tree with recommended options and trade-offs. Use when a plan, design, or idea needs its soft spots found and forced into the open, or when another skill needs the interview protocol.

• handoff

  Compact the current conversation into a handoff document for another agent to pick up.

• harness-tools

  The tool each agent harness (Claude Code, Codex, Grok, agy) uses to start, follow up with, or stop a sub-agent, and to ask the user a structured question. Use when a skill names one of these moves and points here, or when you are not sure which tool your harness has for it.

• housekeeper

  Run a conservative code-project housekeeping pass: one proven low-risk cleanup at a time.

• implement-spec

  Build a whole spec fast: parallel implementer subagents work the ticket frontier, each in its own worktree, merging onto one integration branch; then one review, one verification pass, and one PR or a stack of PRs.

• implement

  Implement one work item end to end, or every ticket of a spec in one run: scope and claim it when trackable, TDD at chosen seams, commit, review, verify, resolve it, and open a PR when asked.

• improve-codebase-architecture

  Scan a codebase for deepening opportunities, present them as a visual HTML report, then grill through whichever one you pick.

• laravel-herd-worktrees

  Create, serve, or remove git worktrees for Laravel projects served by Laravel Herd on macOS. Use when the user needs a branch worktree, a .test URL for one, or per-worktree database isolation.

• panel

  Convene a panel of agent CLIs (grok, claude, codex, agy) on a prompt or a code review, then report one consensus grouped by how much they agree.

• pr

  The pull request body format, built for fast human review. Use when writing or rewriting a PR body.

• prototype

  Build a throwaway prototype to answer a design question. Use when the user wants to sanity-check whether a state model or logic feels right, or explore what a UI should look like.

• prune-tests

  Delete low-signal tests (tests that restate the code, always pass, or break on every refactor) one test area at a time, keeping coverage, and open one PR per area. Every run takes an area no earlier run touched.

• release

  Commit this session's work, bump the version, tag, push, and publish a GitHub release with grouped notes. Use when the user asks to release, cut, tag, or publish a version.

• repository-cleanup

  Audit and clean Git repository state: branches, PRs, stashes, worktrees, and the local databases old worktrees left behind.

• research

  Investigate a question against high-trust primary sources and capture the findings as a cited Markdown file. Use when the user wants a topic researched, docs or package facts gathered, reading legwork delegated, or a wayfinder research ticket resolved.

• resolve-review-comments

  Fix or rebut every review comment on a pull request, human or AI-review bot (Codex, Copilot, Gemini), then reply and resolve each thread. Use when the user wants a PR's review comments addressed, answered, or cleared.

• resolving-merge-conflicts

  Resolve an in-progress git merge or rebase conflict while preserving both sides' intent. Use when a merge/rebase is blocked on conflicts, or the user asks to finish a conflicted merge or rebase.

• retro

  Look back over AI-coding sessions (this one, or recent session logs) and propose improvements to the agent's environment.

• run-after

  Run this session's waiting steps (a backfill after deploy, a command after commit or merge) once their trigger lands. Start the deploy, then run this: it confirms the steps, watches for the trigger, and runs them.

• scratch-cleanup

  Sweep a dusty .scratch/: file research into docs/, remove finished efforts, and report the work still open.

• scratch-status

  Report what is still open across .scratch/ and the next skill to run for each effort. Read-only.

• setup

  Configure this repo for the engineering skills: the markdown tracker layout, triage status vocabulary, and domain doc layout.

• sweep

  Find, prove, and fix one unnoticed problem per cycle, one PR each.

• tdd

  The red → green loop for any stack: seams, what a good test is, and the anti-patterns to refuse. Use when building features test-first or when another skill needs the loop.

• teach

  Teach the user a new skill or concept, within this workspace.

• to-questionnaire

  Turn a decision the user can't fully answer into a Markdown questionnaire for someone else to fill in, async or over a meeting.

• to-spec

  Turn the current conversation into a spec (PRD) saved under .scratch/, by synthesis, not interview. Use when the user wants a spec written from what a discussion, grilling, or wayfinder session settled, or an existing spec revised in place.

• to-tickets

  Break a spec into tracer-bullet vertical-slice tickets with explicit blocking edges, settle the build's open decisions, and save one markdown file per ticket beside the spec. Use when the user asks to break a spec or map into tickets.

• triage-github-pr

  Review and merge a pull request with up to two fix rounds, carrying later findings into a follow-up PR.

• triage

  Move markdown issues and external-PR requests through triage roles: categorise, verify, grill if needed, and write agent-ready briefs.

• ui-design

  The user's UI/UX design rules, each with its reason and a check. Use when building, restyling, or reviewing a page or component, or when the user states a UI preference to record as a rule.

• upgrade-plan

  Plan a library or framework upgrade from the installed version to the latest. Finds what breaks, which local patches and workarounds can go, and what to adopt, sorted into must do, should do, and do later.

• verify

  Exercise a change end to end in the running application (hit the route, run the command, click through the page) and report what actually happened. Use after tests pass, when a change needs proving in the real app, or when another skill's build step ends.

• wait-what

  Stop. That last message did not land. Re-pitch it.

• wayfinder

  Plan work too big for one agent session as a map of decision tickets in markdown under .scratch/, then resolve them one per session, or all in one run, until the route to the destination is clear.

• which-skill

  Not sure which skill to reach for? Name the situation and get routed to one.

• wizard

  Generate an interactive bash wizard that walks a human through steps only they can perform. Use when provisioning infrastructure, creating credentials or CI secrets in a third-party dashboard, or running a one-off migration or cutover. For steps the agent can perform itself, perform them.

• writing-for-agents

  Reference for writing documents an agent consumes. Use when creating or editing a skill, or when modifying AGENTS.md, CLAUDE.md, or a reference doc reached by a pointer.

Subagents (20)

• browser-test-planner

  Turns a start URL and a free-text test prompt into 1–5 independent browser test cases with mechanical checks, marking any step that changes data. Looks at the start page once, read-only. Dispatched by the browser-test skill.

• browser-tester

  Runs ONE read-only browser test case (steps + mechanical checks) with agent-browser and returns a verdict table plus any 4xx/5xx, console, and page errors seen. Never saves data. Dispatched by the browser-test skill; not for tasks that change data.

• browser-writer

  Runs ONE browser test case whose data changes the user confirmed, performing only the named writes, with agent-browser. Returns the same verdict table as browser-tester. Dispatched by the browser-test skill after the user approves a write plan; never for unconfirmed writes.

• chatter-sweeper

  Sweeps one source group of a chatter-scout profile with web searches and returns candidate items as JSON. Search and listing pages only, no deep reading. Dispatched by the chatter-scout skill, one per source group, in parallel.

• chatter-verifier

  Fetches one batch of chatter-scout candidates, confirms each page's date and substance against the topic lens, and returns one verdict per item with the facts it carries. Read-only on the web. Dispatched by the chatter-scout skill, one per batch of about 5, in parallel.

• deepening-scout

  Read-only sweep of a codebase for shallow modules and deepening opportunities, armed with the codebase-design vocabulary. Returns candidates, not file dumps. Dispatched by the improve-codebase-architecture skill.

• dependency-auditor

  Runs the dependency-audit skill in isolation. Detects Composer or npm, runs the outdated and audit commands, and returns the upgrade report. Never applies upgrades. The dependency-audit skill forks into this agent; not for ad-hoc use.

• finding-verifier

  First-pass verification of review findings (PR comments or review-bot issues) against the code. Checks each claim with rg, git, and file reads, and returns one classification line per finding. Read-only. Dispatched by the fix-github-issues, triage-github-pr, and resolve-review-comments skills, one p

• module-designer

  One independent interface design for a module being deepened, under a single design constraint. Preloads the codebase-design and domain-modeling vocabulary. Read-only. Dispatched by the codebase-design skill's design-it-twice reference, several in parallel.

• panelist

  Runs one external agent CLI (claude, codex, grok, agy) headless on a fixed task and distills its output into normalized points. Dispatched by the panel skill, one per CLI, in parallel.

• pr-reviewer

  Independent correctness review of a pull request's diff (bugs, security, data, broken contracts, missing tests), done without seeing the PR's other reviews. Returns one severity-tagged line per finding. Read-only. Dispatched by the triage-github-pr skill, and as code-review's Correctness axis.

• release-notes-collector

  Reads one package's changelog, release notes, and upgrade guide across one span of releases and returns every change as a typed JSON item with its source URL. Web-only, no shell. Dispatched by the upgrade-plan skill, one per release span, in parallel.

• researcher

  Runs the research skill's brief in isolation. Investigates a question against primary sources, writes a cited Markdown findings file, and reports the answer plus the path. The research skill forks into this agent; not for ad-hoc use.

• session-miner

  Mines one batch of AI-coding session transcripts (Claude Code, Codex, Grok, and agy) for corrections, errored commands, repeated setup, and content moments. Read-only over the logs. Dispatched by the retro skill, one per batch.

• smell-sweeper

  One detection pass of the code-smells-audit skill. Sweeps a target list against the one-line smell descriptions of one or more occurrence lenses and returns candidates for verification. Read-only. Dispatched by the code-smells-audit skill for medium and large sweeps.

• spec-reviewer

  Spec axis of the code-review skill. Compares a diff against the originating ticket, PRD, or spec and reports missing, extra, or wrong behaviour. Read-only. Dispatched by the code-review skill with a brief; not for ad-hoc use.

• standards-reviewer

  Standards axis of the code-review skill. Reads a diff and this repo's documented conventions, reports violations and baseline smells. Read-only. Dispatched by the code-review skill with a brief; not for ad-hoc use.

• test-auditor

  Audits one batch of test files against the tdd value gate and returns one Keep/Rewrite/Move/Delete line per test, naming the stronger test behind each Delete. Read-only. Dispatched by the prune-tests skill, one per batch, in parallel.

• ui-reviewer

  Reviews a page, screenshots, or a UI diff against the ui-design rules and returns one line per finding, naming the rule. Read-only. Dispatched by the ui-design skill's Review section.

• upgrade-impact-scout

  Read-only sweep of a codebase for one package upgrade. Inventories local patches, pins, and workarounds for the package, or checks a batch of upstream changes against the code and cites every file:line each one touches. Dispatched by the upgrade-plan skill.

Hooks (5)

• SessionStart

  bash "${CLAUDE_PLUGIN_ROOT}/hooks/bare-worktree-notice.sh"

• PreToolUse

  python3 "${CLAUDE_PLUGIN_ROOT}/hooks/shell_guard.py"

• PreToolUse

  python3 "${CLAUDE_PLUGIN_ROOT}/hooks/memory_write_guard.py"

• PreToolUse

  sh "${CLAUDE_PLUGIN_ROOT}/hooks/terminal-bell.sh"

• Notification

  sh "${CLAUDE_PLUGIN_ROOT}/hooks/terminal-bell.sh"