ll-labs
01Plugin Claude Code
A public marketplace for Louis Lombardi's coding-agent skills.
Plugin Claude Code
A public marketplace for Louis Lombardi's coding-agent skills.
Plugin Claude Code
A small collection of practical coding-agent skills for Claude Code.
Skill Claude CodeCodex
Read-only discussion and consultation mode — answer questions, explain code, weigh trade-offs, and recommend improvements without implementing anything. Use when the user invokes /ask, asks "what does this do?", presents an idea for discussion, requests a recommendation or suggestion, critiques a feature ("I don't…
Skill Claude CodeCodex
Reviews a specific area of the codebase for security, performance & scalability, code quality, or test coverage. Invoke with a path, e.g. /audit apps/backend/auth.
Skill Claude CodeCodex
Shapes a raw, in-your-head idea into a scoped, non-technical concept brief (IDEA.md) through a collaborative one-question-at-a-time interview — the pre-planning step that comes before there is any plan to refine. Use when the user has an unformed idea to flesh out or scope, says "brainstorm", "help me scope/shape this…
Skill Claude CodeCodex
Breaks an existing plan, PRD, or tech spec into small, paste-ready task tickets — one self-contained prompt per task that drops cleanly into the current context window or a fresh one. Use when the user says "breakdown", "break this down", wants a plan/PRD/spec split into actionable tasks, steps, or tickets, or wants…
Skill Claude CodeCodex
Ultra-compressed communication mode. Cuts token usage 75% by dropping filler, articles, and pleasantries while keeping full technical accuracy. Use when user invokes /caveman, says "caveman mode", "talk like caveman", "use caveman", "less tokens", or "be brief".
Skill Claude CodeCodex
Create a git commit for the current work by default, or write/review a commit message when requested with --print. Supports short and detailed modes. Use when the user invokes /commit-this, asks to commit current changes, asks for a commit message, commit title, commit body, conventional commit, or help splitting…
Skill Claude CodeCodex
Create new agent skills with proper structure, progressive disclosure, and bundled resources. Use when user invokes /create-a-skill or wants to create, write, or build a new skill.
Skill Claude CodeCodex
Creates a compressed CONTEXT.md for one target directory and upserts that directory into the root CONTEXT-MAP.md. Use when the user wants targeted codebase context for an app, package, subsystem, or directory, mentions "create-context", asks to refresh a single CONTEXT.md, or wants agents to navigate a large monorepo…
Skill Claude CodeCodex
Create a high-signal GitHub pull request from the current branch. Use when the user invokes /create-pr, asks to create/open a PR, draft a PR description, write PR title/body, or prepare GitHub changes for review.
Skill Claude CodeCodex
Creates or updates PRD.md from CONTEXT.md and/or IDEA.md. Auto-runs grill-me if no context exists. Use when user wants to write a product requirements document, convert grill-me notes into a PRD, or mentions "create-prd".
Skill Claude CodeCodex
Creates or updates TECHSPEC.md from an existing PRD.md. Detects project type from the codebase and adapts sections accordingly. Runs a blocking requirements coverage check before finalising. Use when user wants to write a technical specification, turn a PRD into a tech spec, or mentions "create-tech-spec".
Skill Claude CodeCodex
Grill the user relentlessly about a plan, decision, or idea. Use when the user wants to stress-test their thinking, or uses any 'grill' trigger phrases.
Skill Claude CodeCodex
Compact the current conversation into a handoff document for another agent to pick up.
Skill Claude CodeCodex
Generates a super-compressed CONTEXT-MAP.md birds-eye index of a codebase at the repo root. Use when the user wants a high-level context map, mentions "context map" or "map-context", wants agents to orient in a large repo without burning context, or needs a routing table to existing or future CONTEXT.md files.
Skill Claude CodeCodex
Investigate a question against high-trust primary sources and capture the findings as a Markdown file in the repo. Use when the user wants a topic researched, docs or API facts gathered, or reading legwork delegated to a background agent.
Skill Claude CodeCodex
Review, triage, and optionally address pull request review feedback. Use when the user says "Review PR feedback", asks to handle PR review comments, requested changes, unresolved GitHub PR conversations, inline review comments, or reviewer questions for a PR, branch, repository, or subset of comments.
Skill Claude CodeCodex
Review code changes on the current local git branch and print findings directly in the terminal — never opens, reads, or comments on a pull request. Use when the user says "review-local-branch" or asks to review the current branch, a specific commit, the last commit, the current working diff, or one file's changes on…
Skill Claude CodeCodex
Review a GitHub pull request using the target repo's own review preferences, and post the findings as inline comments directly on the PR. Detects whether this is a first review or a re-review of a PR it already commented on — on re-review it checks whether each of its own prior comments was addressed, replies and…
Skill Claude CodeCodex
Disciplined diagnosis loop for hard bugs and performance regressions. Reproduce → minimise → hypothesise → instrument → fix → regression-test. Use when user invokes /run-diagnose, says "diagnose this" / "debug this", reports a bug, says something is broken/throwing/failing, or describes a performance regression.
Skill Claude CodeCodex
Turn the current conversation into a spec and publish it to the project issue tracker — synthesis of what's already been discussed, not a fresh interview.
Skill Claude CodeCodex
Test-driven development with red-green-refactor loop. Use when user invokes /use-tdd, wants to build features or fix bugs using TDD, mentions "red-green-refactor", wants integration tests, or asks for test-first development.
Skill Claude CodeCodex
Stop. That last message did not land — re-pitch it.