mirror of https://github.com/alirezarezvani/claude-skills.git synced 2026-09-14 23:21:05 +00:00

feat(engineering): add workflow-builder skill for Claude Code workflows

New engineering/workflow-builder plugin: an intake-first skill that designs
and writes deterministic multi-agent workflow .js files for Claude Code's
Workflow tool (CLAUDE_CODE_WORKFLOWS=1, /workflows).

- Opens every session with the intake question set; when the user is vague,
  a stdlib recommendation engine infers and proposes a topology with rationale
  instead of stalling or interrogating in a loop.
- 3 stdlib Python tools: workflow_intake.py (topology + model + budget +
  rationale recommender), validate_workflow.py (.js linter enforcing
  pure-literal-meta / no-non-determinism / no-Node-APIs / parallel-thunk /
  guarded-loop rules), scaffold_workflow.py (5-topology starter generator).
- 3 references (7-8 sources each): API surface, orchestration patterns,
  decision + intake guide. 3 templates + a runnable PR-triage example.
- cs-workflow-architect agent + /cs:workflow-build command.
- Registered in marketplace.json (61 -> 62 plugins).

Conceptually inspired by Ray Amjad's claude-code-workflow-creator; all content
written fresh from the publicly-documented Workflow tool API.

Gates: description PASS, structure PASS (78 lines), review-checklist WARN
(skill vs Workflow tool terminology — official feature name), complexity 85/100.
All scripts pass --help/--sample; all shipped .js validate PASS.

https://claude.ai/code/session_01Q1kXbgMRodzhdTpgbCqVgx

2026-05-27 14:20:53 +00:00

3.5 KiB

Raw Permalink Blame History

name	description
cs-workflow-build	/cs:workflow-build <task-description> — Design and write a deterministic Claude Code workflow (.js). Opens with intake questions, infers-and-proposes a topology when the request is vague, then scaffolds + validates the file. Use when building or running a custom Claude Code workflow.

/cs:workflow-build — Workflow Architect Intake

Command: /cs:workflow-build <task-description>

Designs a deterministic multi-agent workflow for Claude Code's Workflow tool. Always opens with intake; never writes a file before the topology is confirmed.

When to Run

Building a new .claude/workflows/*.js file
Automating a repeatable, multi-step task across fresh-context sub-agents
Deciding whether a task even warrants a workflow (vs. a single agent or a skill)

Step 1 — Intake (always first)

Ask the opening question set. Lead with #1.

What repeatable, multi-step task do you want to automate?
What is the one unit of work a single sub-agent does once?
How many units — a known list, or discovered by looping?
Do later steps need all prior results at once, or can each item flow on its own?
Does any step need structured data back (a verdict, a list, scores)?
Roughly how deep / how many tokens?

Step 2 — If the user is vague, infer and propose (don't loop on questions)

python ../skills/workflow-builder/scripts/workflow_intake.py --task "<their request>" \
  --units unknown --stages unknown --needs-all unknown --structured unknown

Present the result as "here's what I'd build and why": recommended topology (+ runner-up), per-stage model picks, a budget guard, and the rationale for each choice. Then ask only "what should I change?"

Step 3 — Confirm the shape, then scaffold

python ../skills/workflow-builder/scripts/scaffold_workflow.py \
  --topology <fan-out|pipeline|barrier|loop|judge-panel> --name <name> --description "..." \
  > .claude/workflows/<name>.js

Step 4 — Validate before running

python ../skills/workflow-builder/scripts/validate_workflow.py .claude/workflows/<name>.js

Fix every FAIL. WARNs need a one-line justification.

Step 5 — Run

export CLAUDE_CODE_WORKFLOWS=1   # the feature is off by default
# Save under .claude/workflows/, then launch + monitor via /workflows.
# P = pause/resume, X = skip a sub-agent. Failed agents retry automatically.

The Hard Rules (validator enforces)

meta is a pure literal and the first statement — no variables, spreads, template strings, or calls.
No Date.now(), Math.random(), or argless new Date() — they break resume.
No filesystem / Node APIs in the orchestrator — that work goes inside agent().
parallel() takes thunks (() => agent(...)); default to pipeline() unless a stage needs the whole prior set.
Guard every open-ended loop with a counter or budget.remaining().
results.filter(Boolean) before using parallel/pipeline output.

Output Format

# Workflow Build: <name>
## The Decision
[intake | recommend | scaffold | validate | run]
## Recommended Topology
[fan-out | pipeline | barrier | loop | judge-panel] — why
## Model Plan
[per-stage model + reason]
## Validator Verdict
🟢 PASS | 🟡 WARN (justified) | 🔴 FAIL (with line numbers)
## Next Steps
[3 concrete actions]

Agent: cs-workflow-architect
Skill: workflow-builder
Adjacent: /cs:write-a-skill (authoring skills, not workflows)

Version: 1.0.0

3.5 KiB Raw Permalink Blame History