by-carlos
UserClaude Code plugin marketplace — run big projects as many small, resumable sessions instead of one huge one.
Categories
Indexed Skills (26)
staged-rollout
Use when a build is too big for one session and must be run as many small, resumable sessions — a staged or milestone rollout with cross-session progress tracking, a persistent plan that must not drift, and stop/resume freely over hours or days. Covers decomposing a large project into a `.plan/` folder of dependency-ordered stages with an evidence ledger, and executing one stage per fresh session. NOT for single-session tasks, quick fixes, or work of roughly three sessions or less — the scaffold has a floor cost that only pays off on genuinely large, decomposable builds.
project-summary
Gives a full-state overview of a project's Daikenja ledger, written for someone with no prior context. Read-only; writes nothing. Accepts an optional project key -- `/daikenja:project-summary <key>` reads that project from anywhere.
thread
Reads a Slack or email thread, summarizes what is being asked and by whom, then collects context from the user before any reply is drafted. Use when the user pastes a thread link or a block of chat history, or says "help me answer this". It gathers only and never writes the reply.
doc-review
Reviews a document against a predetermined checklist: clarity, undefined terms, unenforced rules, undated decisions, missing owners, contradictions, and obvious stale content. Use when the user says "review this doc", "review this page", "check this before I publish", or pastes a document and asks for a check before sharing it. Read-only -- it reports findings, it never edits or rewrites the document.
meeting-review
Turns a meeting transcript into proposed ledger entries: decisions made, action items with owners and dates, and questions left unresolved. Use when the user pastes a transcript or asks what came out of a call. It classifies only and writes nothing itself.
preflight
Challenges a draft before it goes out and hands back a revised version plus the facts only you can supply -- the six substance checks, then a set of reviewer personas whose wording problems it fixes. Never sends anything.
project-catchup
Reports what changed in a project's Daikenja ledger since the user last checked, then advances the checkpoint on approval. The only skill that writes last_checkpoint in daikenja.yaml; it never touches ledger content. Accepts an optional project key -- `/daikenja:project-catchup <key>` reads that project from anywhere.
project-decisions
Looks up what was decided about a specific topic in a project's Daikenja ledger, including its supersession history and what it is blocked by or contradicts. Read-only; writes nothing. Accepts an optional project key -- `/daikenja:project-decisions <key>` reads that project from anywhere.
project-gaps
Audits a project's Daikenja ledger for open items with no owner or that have sat too long; only Open items are in scope. Read-only; writes nothing. Accepts an optional project key -- `/daikenja:project-gaps <key>` reads that project from anywhere.
project-log
Records decisions, open items and sources in a project's ledger. Use when the user says "log this", "record this decision", "add this to the ledger", or asks for what was settled to be written down.
remember-persona
Records what the user says about a person they write to, so later messages are written for that reader. Use when the user says "remember that S challenges every technical claim", "log this persona", or describes a recipient while drafting and wants that kept.
self-review
Reviews how the user themselves handled a thread they took part in, with private, direct, evidence-backed coaching on their own moves. Findings are about the user alone, and nothing is written, logged or sent.
project-list
Lists every project registered in daikenja.yaml, says which one the current directory resolves to, and reports whether each project's ledger actually exists. Read-only; it never repairs what it reports.
project-sources
Reports which of a project's tracked sources moved since they were last read, by comparing each source's stored last-modified value against what its system reports now. Read-only against the ledger; recording a re-read goes through /daikenja:project-log. Accepts an optional project key -- `/daikenja:project-sources <key>` reads that project from anywhere.
learn-voice
Derives your writing-style.md from writing samples you supply, backing up whatever the file held first. It records style rules only, never facts about people or projects.
compose
Rewrites or writes a message so it stays clear, professional and easy to read for all audiences without changing the ask, stance or confidence level.
setup-user
One-time and re-runnable personal setup for Daikenja: creates your configuration, captures your profile, seeds the persona, writing-style and glossary files, and applies any pending version upgrade.
setup-project
Registers the project you are in with Daikenja, sets its per-project settings, and optionally seeds its ledger from sources the project already has.
tensei
Installs, updates, or removes the Tensei conversational persona (~/.claude/daikenja/tensei.md) and its CLAUDE.md import line, diffing an existing copy and backing up anything it would overwrite.
judgement
Checks a thread or a document against the project ledger and general knowledge and hands back a verdict: what is being claimed or asked, which parts a recorded decision settles, which parts are plain fact, and what is presented as settled without being established. Use when the user says "is this right?", "check this against the ledger", "give me a judgement on this thread", or asks for a quick "AI review summary" of a page or pull request to share with the team. Read-only -- it never drafts a reply, never edits the document and never writes the ledger.
digest
Groups a list of incoming messages by project and enriches each group with that project's open ledger items, for a scheduled digest. Read-only; writes nothing. Takes the item list on standard input and never fetches anything itself.
debloat-context
Use when per-session context overhead needs auditing or reducing — large CLAUDE.md files, long skill listings, memory indexes, or third-party advice to disable Claude Code features — and a measured, step-gated cleanup is wanted instead of blind switch-flipping.
summary
Emit a fixed, skimmable summary format — Context, numbered findings, optional Conclusion, optional Notes. Use ONLY on explicit invocation via the /summary slash command, or explicit phrases like "standard summary," "my summary format," or "give me a summary in your usual format." Do NOT use for incidental or generic requests to "summarize" something — those should get a normal, freeform answer. This skill overrides the user's default brevity rules while active.
triage-issues
Triage GitHub issues across one or more repos and rank them into a Projects (v2) board queue — dedup/consolidate, fix hygiene, and set Status/Priority/Size/Effort so `/work-issue next` can burn the queue down. Runs incrementally by default (deep-reads only untriaged issues); pass `--full` for an exhaustive sweep. Use for backlog triage, deduplication, or board grooming.
work-issue
Work a GitHub issue end-to-end — read the thread, gate on scope, branch, implement with conventional commits, open a PR that closes the issue, and squash-merge after explicit confirmation. Use when asked to work, fix, or implement a GitHub issue by number.
staged-rollout
Run a large build as many small, resumable sessions by decomposing it into a `.plan/` folder of dependency-ordered stages with an evidence ledger, then executing one stage per fresh session. Use when the user says "this is too big for one session", "plan a staged rollout", or describes a multi-day build that needs cross-session progress tracking.
Bio shown is the top-scored skill's repo description as a fallback — real GitHub bios land in a future update.