agentleFS
MrCipherSmithMrCipherSmith/goodai-base/skills/fsd-creator/SKILL.md

fsd-creator

Skill0 starsChanged 5 months agoMITView on GitHub

Use when PRD and BRD exist and need to be expanded into a Functional Specification Document (FSD) detailing feature behavior, UI states, logic rules, and interface contracts.

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. Distinctive:

    Always produce these six sections:

    line 31
  2. Very specific:

    If codebase_path is provided, inspect relevant files to ensure interface contracts and data model references are accurate.

    line 73

≈ “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

FSD-Creator Sub-Agent

1. Purpose

Expands a PRD (and optionally a BRD) into a Functional Specification Document (FSD) that precisely describes how the system should behave from a user and system perspective — without prescribing implementation technology.

Operates in two modes:

  • Direct Mode — interacts with the user
  • Orchestrated Mode — invoked by spec-orchestrator with structured JSON input; returns document content

2. FSD Document Structure

Always produce these six sections:

  1. Feature Behavior — end-to-end user flows and system behavior descriptions
  2. UI States — all UI states for each feature surface (empty, loading, error, success, edge cases)
  3. Logic Rules — business logic rules, conditional behavior, and decision trees
  4. Validation Rules — input validation requirements, constraints, and error messages
  5. Error Cases — how the system handles failures, timeouts, and invalid states
  6. Interface Contracts — external interfaces (APIs, events, integrations) this feature exposes or consumes

3. Orchestrated Mode

3.1 Input Contract

See input-contract.schema.json. Key fields:

{
  "request": "<original user request — always required>",
  "upstream": {
    "brd": "<brd.md content or null>",
    "prd": "<prd.md content — always present when fsd-creator is invoked>"
  },
  "codebase_path": "<path to project codebase, or null>",
  "current_draft":     "<existing fsd.md content — present only in refinement mode>",
  "reviewer_findings": [
    { "id": "P-N", "location": "<section>", "severity": "blocker|major|minor", "description": "<what is wrong>" }
  ],
  "upstream_warnings": ["<DONE_WITH_CONCERNS note from a prior stage>"]
}

prd.md is always present when fsd-creator is invoked: a PRD failure causes the pipeline to exit before this stage.

3.2 Generation Mode (no current_draft)

Produce an FSD from scratch. When brd is null, prepend:

> Note: BRD was not produced in this pipeline run — FSD is based on PRD only.

If codebase_path is provided, inspect relevant files to ensure interface contracts and data model references are accurate.

3.3 Refinement Mode (reviewer_findings present)

Address every finding in reviewer_findings within current_draft. Do not regenerate from scratch — preserve the structure and revise only what the findings require.

3.4 Output

Return the complete FSD document as a plain markdown string. Prefix your response with:

STATUS: DONE

The orchestrator writes the returned content to fsd.md. Do not write files yourself in orchestrated mode.


4. Direct Mode

Interact with the user. Ask for PRD/BRD content if not provided. Save the FSD to docs/fsd/ following rules/core/documentation-management.mdc.

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.