agentleFS
patrickbdevaneypatrickbdevaney/hermes-max/skills/workflow-spec-driven/SKILL.md

workflow-spec-driven

Skill1 starsChanged 6 months agoMITView on GitHub

For a GENUINELY complex, multi-component task, produce durable versioned artifacts before implementing — SPEC.md (what + why), PLAN.md (how), TASKS.md (decomposed checklist) — then work through TASKS.md item by item. Use this ONLY for real multi-component work; it adds rigidity and overhead, so do NOT use it for small fixes or single-file changes.

The file

We split this skill into the rules it gives an agent

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

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈1

    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
  1. Very specific:

    Create these in the project (e.g.

    line 25
  2. Distinctive:

    When reality diverges from SPEC/PLAN, update the artifact — they're durable and versioned (commit them), not write-once.

    line 38

≈ “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

workflow-spec-driven

<!-- TRIGGERS WHEN: a task spans multiple components/files and benefits from an up-front spec; NOT for small fixes -->

GitHub Spec-Kit discipline: for complex work, write the contract down first so the implementation has a stable target and the decomposition is explicit and reviewable.

Honest framing (from the research): this adds rigidity and up-front cost. Use it for genuinely complex, multi-component tasks (a new subsystem, a cross-cutting change, a multi-stage feature). For a small fix, a single-file change, or a mechanical edit, skip it — go straight to the change (and at most [[workflow-plan]]).

The artifacts (write before implementing)

Create these in the project (e.g. under specs/<task>/):

  1. SPEC.md — what and why: the goal, the user-visible behavior, constraints, explicit non-goals, and acceptance criteria. No "how".
  2. PLAN.md — how: the design, the components touched, the sequence, risks and trade-offs. (Pairs with [[workflow-plan]].)
  3. TASKS.md — the decomposed, checkable checklist: each item is one independently verifiable change with its own done-condition (a test, a verify-green, a checkpoint).

Working the spec

  • Implement by walking TASKS.md top to bottom; check items off as each passes its done-condition (verify green + checkpoint per [[workflow-task-finish]]).
  • When reality diverges from SPEC/PLAN, update the artifact — they're durable and versioned (commit them), not write-once. A drifted spec is worse than none.
  • The artifacts are also the input to the regression eval and the trajectory store — a clear SPEC makes a failure localizable later.

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.