agentleFS
ruverdruverd/bulma/skills/bulma-developer/SKILL.md

bulma-developer

Skill1 starsChanged 49 days agoMITView on GitHub

Graph: deliver a ticket or a local goal via grill -> spec -> tickets -> TDD (bulma-feature-delivery), then MERGEABLE + bot_review + bulma-qa over the bus. Use when /developer, /bulma-developer, a ticket, or a QA_RESULT FAIL+PR_BUG. ASK the user only as a last resort.

The file

We split this skill into the rules it gives an agent

1 rule. About 0 are common, 0 are distinctive, and 1 is very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈0

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 1

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  • could be automatic: a setting could enforce it
  1. Very specific:

    Never merge. Stay Draft until QA PASS, then gh pr ready.

    line 26

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Bulma developer (graph)

You are the orchestrator. Chat: bulma-memory. Apply Kevin for human-facing communication. You do not implement product code. Fd coder / fix node does.

REQUIRED (load now):

  • ARGS.md (ticket vs goal vs resume)
  • GRAPH.md
  • STATE.schema.md
  • bulma-host
  • bulma-bus PROTOCOL.md · DISK.md
  • bulma-memory
  • delivery voice + policy: ../bulma-feature-delivery/VOICE.md · ../bulma-feature-delivery/DECISION_POLICY.md · ../bulma-feature-delivery/PSTACK.md

Never merge. Stay Draft until QA PASS, then gh pr ready. Ready is not merge. Worktree and branch rules: JOBS.md §Worktree.

Start

Parse $ARGUMENTS with ARGS.md before anything else.

Resume (resume / same tracker id with existing STATE): replay invariants (ARGS.md), continue at the saved node. Skip finished only when invariants match. The current message is the answer if waiting_user. Do not re-init STATE. Do not re-grill settled decisions unless an AC line contradicts them.

Goal or ticket (free text, or tracker id/URL, no live STATE for that id):

  1. Resolve $BULMA_ROOT (../bulma-bus/DISK.md).
  2. Init .bulma-developer/STATE.md from templates/STATE.md.
  3. admit (nodes/admit.md / JOBS.md). Second job while another owns main → worktree + general-purpose worker. Do not abort QA.
  4. Foreground: QA_RESULT FAIL + PR_BUG → mode fix; else delivery.
  5. Walk GRAPH. Spawn nodes, never other graphs as subagents.
  6. Outbound QA → enqueue-or-start (JOBS.md), never spawn bulma_qa.

/bulma-developer <goal>, /bulma-developer ABC-123, and /bulma-developer resume are first-class.

Modes

InputPath
Goal / ticket / feature / bug / choredeliver = run bulma-feature-delivery until CI green
QA_RESULT FAIL + PR_BUGfix = existing PR branch only
Stacked task (sub-issues, several ids, "stacked")stack = one worker per child running Bulma, PRs on main, orchestrator fixes any red CI (STACKED.md)

Delivery spine (inside fd): grill → spec → tickets → implement(TDD). Bugs go through diagnose first. TDD iron law. Thermo before PR.

Extra gates (after delivery / after fix)

CI green AND mergeability = MERGEABLE. See references/GATES.md. Then bot_review (nodes/bot_review.md). Then envelope QA_REQUEST. See references/QA_HANDOFF.md.

Resume

Full rules: ARGS.md. STATE + delivery STATE/HANDOFF + bus stack. Do not restart delivery if already green.

/bulma-developer the notification inbox on the dashboard
/bulma-developer ABC-123
/bulma-developer ABC-123: extra note
/bulma-developer resume
/bulma-developer resume: the answer is B

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.