agentleFS
raniellimontagnaraniellimontagna/agent-toolkit/skills/workflow/ariadne-prd/SKILL.md

ariadne-prd

Skill3 starsChanged 32 days agoMITView on GitHub

Use when turning product requirements, plans, or an existing backlog into an Ariadne schema version 1 PRD.

The file

We split this skill into the rules it gives an agent

6 rules. About 0 are common, 3 are distinctive, and 3 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈3

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 3

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Create the smallest executable backlog that preserves the user's intent.

    line 8
  2. Very specific:

    Read applicable repository instructions and inspect existing code, tests, .ariadne/prd.json, and .ariadne/config.json when present.

    line 15
  3. Very specific:

    Write valid JSON directly to .ariadne/prd.json at the repository root.

    line 23
  4. Very specific:

    Use agent-toolkit ariadne init --check <command> to establish them.

    line 26
  5. Distinctive:

    Every story needs non-empty, observable acceptance criteria.

    line 54
  6. Distinctive:

    No implementation, status transition, commit, or push is performed while authoring the PRD.

    line 68

≈ “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

Ariadne PRD

Create the smallest executable backlog that preserves the user's intent. Ask questions only for requirements that remain unresolved after inspecting the request and repository; do not re-ask answered questions or invent product-critical behavior.

Workflow

  1. Read applicable repository instructions and inspect existing code, tests, .ariadne/prd.json, and .ariadne/config.json when present.
  2. Resolve only material unknowns with focused questions. Do not write an affected story until its material requirement is resolved. If a reversible, low-risk assumption is enough, state it and continue.
  3. Split work into independently testable stories that one runtime agent can implement and verify in one Ariadne iteration. Separate unrelated outcomes, migrations, and cross-subsystem changes.
  4. Write valid JSON directly to .ariadne/prd.json at the repository root. Preserve completed stories and stable IDs when revising an existing PRD.
  5. Confirm that .ariadne/config.json names the real repository quality checks. Use agent-toolkit ariadne init --check <command> to establish them; never add unsupported fields to the PRD schema.

Required schema

{
  "schemaVersion": 1,
  "project": "project-name",
  "branchName": "current-branch",
  "description": "Measurable project outcome",
  "userStories": [
    {
      "id": "US-001",
      "title": "Small independently verifiable outcome",
      "description": "What changes and why",
      "acceptanceCriteria": [
        "Observable pass/fail behavior",
        "Relevant automated test or quality evidence passes"
      ],
      "priority": 1,
      "status": "pending",
      "attempts": 0
    }
  ]
}

Every story needs non-empty, observable acceptance criteria. Include relevant testing, lint, typecheck, build, security, compatibility, or migration evidence without naming commands that the repository does not provide. Priorities are positive integers; lower values run first. New stories start pending with zero attempts.

Quality gate

  • .ariadne/prd.json uses only the schema fields above and schema version 1.
  • IDs are unique, branch and project are correct, and stories fit one iteration.
  • .ariadne/config.json.qualityChecks contains at least one real command before a normal run.
  • Story acceptance evidence identifies which configured checks prove the relevant behavior; repository-wide checks remain in config, not PRD fields.
  • No implementation, status transition, commit, or push is performed while authoring the PRD.

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.