cj-lab

Plugin cj-lab from marketplace cj-stack in cpeaustriajc/cj-stack. Commands read from the repo's manifest — 18 days ago.

Skills (20)

• architect

  This skill should be used when the user types /architect with a task: sketch types, signatures, and module structure before code, then stay in the loop while implementation fills in. Use for 'architect this', 'design this', or non-trivial work where jumping to code would lock in the wrong shape.

• blast-radius

  This skill should be used when the user types /blast-radius on a change: find what a change could break somewhere else before it ships, beyond the diff, and prove the one fact it's safe because of by running real code instead of writing it up. Use for 'blast radius of X', 'what could this break', or

• bro

  This skill should be used when the user types /bro. Stop. That last message did not land: re-pitch it. Restate the last message in plain human language, with no jargon, the context the user was missing first, and one picture.

• bug-hunt

  This skill should be used when the user says "bug hunt", "deep code review", "audit the codebase", "hunt for bugs until clean", "run N rounds of review" or "find and fix every bug". Three parallel hunters search the whole scope for real bugs, fix each one test-first in files they own, and search aga

• cleanup

  This skill should be used when the user says "cleanup", "clean up after yourself", "tear down the dev server", "archive this session" or "clean up stale branches and worktrees". It removes what a session left behind (servers, containers, branches, stashes, evidence) once each item is proven safe, na

• drive

  This skill should be used when the user says "drive this", "loop until it passes", "keep going until green", "run it overnight" or "hillclimb", hands over a ticket to finish unattended, or before writing a goal or Stop hook. It decides whether the work should loop, writes the exit check first, freez

• how

  This skill should be used when the user types /how with a question: \"how does X work\", code walkthroughs before changing something, and placement / ownership / layering questions (\"where should this live\", \"which package owns this\", \"is this the right layer\"). Explains subsystem architecture

• implement

  This skill should be used when the user types "/implement" with a ticket or a spec. It builds one ready ticket without supervision: it gets the ticket from the tracker, trusts its answers, builds it with the feature playbook, reviews after the commit and opens a PR that closes it. With a spec or "or

• ketchup

  This skill should be used when the user types /ketchup. Catch the user up on what changed in this chat since their last message, in a few plain sentences and small visuals. Give each action item just enough context to make the call. Use for /ketchup.

• launch-video

  This skill should be used when the user asks for a trailer, teaser, launch video, promo, reel, short, TikTok, ad, "something for social media", a voiceover, thumbnails or launch images, or captions for a video, or says a promo feels flat, "meh" or robotic. It makes a narrated launch video from a liv

• microinteraction

  This skill should be used when the user types "/microinteraction" with an area of a web UI. It searches the area for places that don't animate but should, rejects everything that shouldn't, plays each proposal inline on a stand-in copy of the element with its exact values, and waits for the owner to

• no-comments

  This skill should be used when the user types /no-comments with a folder of old code: spawn the sweeper, fix accepted findings, land comment deletions, suppression removals and reshapes as separate commits with the same test results, offer encodings for claimed constraints, and hand each MUST KILL t

• principles

  This skill should be used when the user names an engineering principle, such as "apply prove it works", "subtract before you add" or "attack the premise", and before a design, test, debugging or measurement decision. It indexes 20 principles; Claude reads the one it applies and cites it with the dec

• recall

  This skill should be used when the user says "recall", "where was I", "catch me up", "what did we do last week", "what's in flight" or "pick up where we left off", or at the start of a session that resumes work. It writes a short brief of the recent work in this project from past transcripts and the

• scout

  This skill should be used when the user says "find me an idea", "what should I build", "is there a SaaS in this", "scout this niche", or asks what is still worth building when apps are cheap to copy. It runs afternoon methods for finding demand, filters candidates for what a clone can't take, and ha

• self-verify

  This skill should be used when the user says "set up verification", "make agents verify their work", "verify skill", "agents keep claiming done without proof", "we can't review every PR", "set up PR verification" or "audit the verify skill". It sets up and maintains the project's own verify skill an

• teach

  This skill should be used when the user types /teach with a subject: explain a body of work plainly so a person actually understands it. Runs the `how` and `why` skills and weaves what they find into one clear explanation. Use for 'teach me this', 'help me really understand X', 'explain this change

• triage

  This skill should be used when the user says "triage #42", "triage the issues", "what needs my attention", "is this ready for an agent", "can an agent take this", "move #42 to ready-for-agent" or "write an agent brief", and before an agent starts to build an issue or take over a PR. It checks the is

• unslop

  This skill should be used when the user types /unslop with text. Cut AI tells from any writing.

• why

  This skill should be used when the user asks "why is this code like this", "why was X set to 5", "where did this number come from", "can I delete this guard", "is this check still needed" or "what's the history behind this", and before removing defensive code nobody explains. It digs through git his

Similar plugins