task-board-init — file-based Kanban scaffolder

opus

Tip

One command, complete task system. Point the skill at any repo and it spawns parallel analysis subagents, confirms the findings with you, then writes a fully parametrized file-based Kanban: curator agent, on-demand dashboard skill, lifecycle rule, and the complete .claude/features/** folder tree. After it runs, the repo is self-contained — no further dependency on this generator.

Quick reference

Item	Value
Skill command	`/brewtools:task-board-init`
Arguments	`[target repo path \| empty = cwd] [free-text directive]` — e.g. `'also dedupe rules'`, `'skip module split'`
Emits	`.claude/agents/task-tracker.md`, `.claude/skills/task-board/SKILL.md`, `.claude/rules/tasks.md`, `.claude/features/**`
Spawns	Parallel analysis subagents (architect + Explore) + multi-agent doc sweep — one bounded unit each
Optional	Gated `CLAUDE.md` optimization pass — propose-only, opt-in during confirmation
Model	opus
Commits?	Never — committing is a user / manager action

When to use

Scenario	Command
New repo needs file-based task tracking	`/brewtools:task-board-init`
Standardize an existing Kanban convention across repos	`/brewtools:task-board-init /path/to/repo`
Migrate scattered TODO / backlog docs into a structured board	`/brewtools:task-board-init` — sweep subagents migrate legacy docs automatically

Example

# Deploy into the current repo
/brewtools:task-board-init

# Deploy into another repo
/brewtools:task-board-init /path/to/some-repo

# Deploy, then steer the optional CLAUDE.md pass with a free-text directive
/brewtools:task-board-init /path/to/some-repo also dedupe rules, skip module split

After completion the repo gains /task-board (view/add/move tasks) and the task-tracker agent auto-runs at the start of every task.

How it works

Multi-agent repo analysis
Parallel subagents (architect + Explore) derive: domain id-segments, source-dir exclusions the curator must never touch, release style (vX.Y.Z tag / commit SHA / none), doc language, and an inventory of existing task docs. Findings are confirmed via AskUserQuestion — generation does not start until the user approves. If analysis yields no domains, the skill asks the user to name at least one (falls back to a single CORE domain).
Generate task-tracker agent
Writes .claude/agents/task-tracker.md — the board curator. It owns the Kanban: creates, moves, and closes tasks; grooms the backlog; keeps board.md in sync with the folder state. The agent writes only under .claude/features/** and never touches source directories. Release style shapes the closing-marker wording (vX.Y.Z tag + commit SHA / bare SHA / date / no tag).
Generate task-board skill
Writes .claude/skills/task-board/SKILL.md — the on-demand dashboard. Supports flows: view, add, move, backlog, groom. Non-trivial and bulk passes are delegated to the task-tracker agent. This skill is the everyday interface; the agent is the heavy-lifting backend.
Generate rule + scaffold + doc sweep
Writes .claude/rules/tasks.md (paths-scoped to .claude/features/**) — lifecycle rules, id convention, required frontmatter, grooming cadence, and the run-at-task-start rule that spawns task-tracker as an isolated subagent at the beginning of every task. This rule lives only in tasks.md — board generation never edits the target’s CLAUDE.md. Then scaffolds the folder tree and launches parallel sweep subagents over the discovered doc inventory to migrate legacy backlog / feature docs into the board (dedup, move done items into closed/, author initial board.md counts). If no legacy docs were found, the sweep is skipped entirely.
Optional CLAUDE.md optimization (gated)
Runs only if you opted in during the Step 1 confirmation. Propose-only: it reports current-vs-target line count, then asks before every change — extracting local-only content to CLAUDE.local.md, decomposing an over-budget file into nested module CLAUDE.md files (loaded on demand, unlike eager @path imports), deduping rules, then handing the touched files to text-optimize for token compression. Decline it and nothing is written. Pass a free-text directive in the argument (“also dedupe rules”, “skip module split”, “report only”) to steer this phase.

Delegation

Warning

A big task handed to one agent = an agent gone for an hour. You cannot observe it, you cannot correct it, and it usually drifts off-target. Both spawn points — the Step 1 repo analysis and the doc sweep — give one subagent ONE bounded unit: one doc group, ~5 files, ~10 steps. Bigger work is split into N tasks, all spawned in a single message.

Every spawn prompt carries six fields. A bare one-line task is never enough:

Field	Content
GOAL	deploy the board into the target repo; this pass fills it from pre-existing task docs
ROLE	own the listed docs — no invented tasks, no source-dir edits, no `CLAUDE.md`
SCOPE	write only under `.claude/features/**`; out — exclusions, `CLAUDE.md`, `.claude/agents`, `.claude/skills`
CONTEXT	confirmed domains / release style / language, plus the already-written skeleton and templates; siblings sweep other groups now
CONSUMER	verification counts what landed, and the installed `task-tracker` agent reads those files from then on
DONE	files under `closed/` + `backlog/` plus a manifest — a no-op sweep must say so explicitly

The CONSUMER field is load-bearing here: an id or status folder that deviates from TASK_TEMPLATE.md makes the task invisible to the task-tracker agent forever after.

Note

The same six-field brief ships into your repo. The generated .claude/skills/task-board/SKILL.md carries its own Delegation section with the identical GOAL / ROLE / SCOPE / CONTEXT / CONSUMER / DONE contract, so bulk groom and transition passes delegated to task-tracker are briefed the same way long after this generator is gone.

What gets deployed

Artifact	Path	Role
Curator agent	`.claude/agents/task-tracker.md`	Owns the board — create / move / close tasks, groom backlog, keep `board.md` canonical. Writes only `.claude/features/**`.
Dashboard skill	`.claude/skills/task-board/SKILL.md`	On-demand `/task-board` — view / add / move / backlog / groom; delegates bulk to the agent.
Paths-scoped rule	`.claude/rules/tasks.md`	Lifecycle, id convention, required FM, grooming — plus “run `task-tracker` at the start of any task”. Auto-loads on `.claude/features/**`.
Board file	`.claude/features/board.md`	The Kanban board — counts and task table, authored by the doc sweep.
Control files	`.claude/features/{TRACKER,TASK_TEMPLATE,INDEX}.md` + `backlog/README.md`	Tracker log, task template, index, backlog guide.
Folders	`.claude/features/{backlog,todo,progress,closed,specs}/`	Status folders — folder name = task status.

ID convention

Task IDs follow UPPER-KEBAB: <PREFIX>-<DOMAIN>-<SLUG>.

Prefix	Use
`T-`	Feature / product task
`BUG-`	Defect
`M-`	Maintenance / refactor / tech-debt
`EPIC-`	Umbrella over several tasks

<DOMAIN> is the per-repo first kebab id-segment, discovered and confirmed in Step 1 (e.g. a frontend project might use UI, API, BUILD).

Internals: placeholder map + reference files

Placeholder map — derived from confirmed Step 1 findings and substituted into every template before writing:

Placeholder	Derivation
`{{DOMAINS}}`	Confirmed domain id-segment list, comma-separated
`{{FIRST_DOMAIN}}`	`DOMAINS[0]`
`{{EXCLUSIONS}}`	Confirmed source-dir exclusion list
`{{REPO_NAME}}`	`basename` of `TARGET`
`{{LANG}}`	Confirmed doc language
`{{TODAY}}`	ISO date (`YYYY-MM-DD`)
`{{CLOSE_MARKER}}` / `{{CLOSE_MARKER_SHORT}}`	Derived from `RELEASE_STYLE`: `vtag` → `vX.Y.Z tag + commit SHA` / `vX.Y.Z tag`; `sha` → `commit SHA`; `none` → `date / no tag / superseded / cancelled` / `no tag`

Reference files (loaded from ${CLAUDE_SKILL_DIR}/references/):

File	Purpose
`01-analysis.md`	Step 1 analysis prompts + AskUserQuestion confirmation contract
`02-task-tracker-agent.md`	`task-tracker` agent template (placeholders)
`03-task-board-skill.md`	`task-board` skill template
`04-tasks-rule.md`	`tasks.md` rule template (includes run-at-start rule)
`05-features-templates.md`	`.claude/features/**` file templates
`06-doc-sweep.md`	Multi-agent doc-sweep procedure
`07-claude-md-optimize.md`	Optional gated CLAUDE.md optimization procedure (P5.5)

📋

Updating plugins

Use /brewtools:plugin-update to check and update the brewcode plugin suite in one command. See the FAQ for details.