agentleFS
hardness1020hardness1020/VibeFlow/.claude/skills/create-feature-spec/SKILL.md

create-feature-spec

Skill8 starsChanged 8 months agoMITView on GitHub

Create Feature Specs for Stage E of the VibeFlow docs-first workflow

The file

We split this skill into the rules it gives an agent

2 rules. About 0 are common, 0 are distinctive, and 2 are 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.

  • 2

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Create Feature Specs for Stage E of the VibeFlow docs-first workflow.

    line 17
  2. Very specific:

    Document design changes (UI/API/schema)

    line 24

≈ “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

create-feature-spec

Create Feature Specs for Stage E of the VibeFlow docs-first workflow.

Purpose

This skill creates Feature Spec documents that:

  • Define the API design contract (exact function/class signatures)
  • Specify testable acceptance criteria
  • Document design changes (UI/API/schema)
  • Include test and evaluation plan
  • Reference discovery findings and specs

Workflow

Stage E: Feature Spec
    │
    ├── Define API Design (exact signatures)
    ├── Write testable acceptance criteria
    ├── Document design changes (UI/API/schema)
    ├── Create test and evaluation plan
    ├── Map telemetry and metrics
    └── Reference discovery findings and specs

Usage

Create Feature Spec

/create-feature-spec <ID> <slug>

Creates docs/features/ft-<ID>-<slug>.md with required sections.

Example:

/create-feature-spec 030 anti-hallucination

Validate Feature Spec

/create-feature-spec validate <ID>

Validates that the feature spec has all required sections.

Update Schedule

/create-feature-spec schedule

Updates docs/features/schedule.md with feature status.

Document Requirements

Required Sections

For All Features:

  • Header (ID, File, Owner, TECH-SPECs with versions)
  • Architecture Conformance
  • API Design (exact signatures)
  • Acceptance Criteria
  • Design Changes
  • Test & Eval Plan
  • Telemetry & Metrics
  • Edge Cases & Risks

For Medium/Large Features (Additional):

  • Stage B Discovery Findings
    • Test Impact Analysis
    • Existing Implementation Analysis
    • Dependency & Side Effect Mapping

API Design Section (Critical)

The API Design section defines the contract for Stage F (test writing):

## API Design

### ServiceName.method_name()
- **Signature:** `method_name(param: Type, param2: Type = default) -> ReturnType`
- **Purpose:** [Brief description]
- **Parameters:**
  - `param`: [Description]
  - `param2`: [Description, default value]
- **Returns:** [Description of return type and structure]

### API Endpoint: POST /api/v1/endpoint
- **Method:** POST
- **Path:** `/api/v1/endpoint`
- **Request Body:**
  ```json
  {
    "field": "type"
  }
  • Response Body:
    {
      "result": "type"
    }

Why This Matters:

  • Stage F creates implementation stubs from these signatures
  • Tests are written using these exact function names
  • Any changes require updating specs first (Stage G.1)

Acceptance Criteria

Format as testable checklist or Gherkin:

## Acceptance Criteria

- [ ] User can [action] with [condition] resulting in [outcome]
- [ ] System returns [response] when [condition]
- [ ] Error [X] is shown when [condition]

Or Gherkin:

Given [context]
When [action]
Then [expected outcome]

Validation

  • scripts/validate_feature.py — Validate feature spec structure and API Design

References

See assets/:

  • feature-template.md — Complete feature spec template

See references/:

  • api-design-guide.md — API Design section guidance

Context7 Library Research

Follow the workflow in .claude/rules/context7-research.md to fetch current library documentation.

What to look for: exact method signatures, error types, configuration options, parameter defaults.

Where to incorporate findings:

  • API Design — use verified signatures and types from library docs, not assumed ones
  • Acceptance Criteria — reference actual error types and edge cases from docs
  • Edge Cases & Risks — document known limitations and version-specific behaviors

Manifest Update

After completing Stage E, update docs/workflow-state.yaml:

  • Set stage: E
  • Set docs.feature: docs/features/ft-<ID>-<slug>.md

Checkpoint #2 (after Stage E):

  • Set checkpoint: 2 after passing validation
  • Criteria: Feature spec exists with all required sections, API Design section has exact signatures, acceptance criteria are testable

Git Commit

After completing this stage, ask the user for permission before committing:

git add docs/features/ft-<ID>-<slug>.md docs/workflow-state.yaml
git commit -m "feat(feature): define <slug> feature spec (#ft-<ID>)"

Replace <ID> and <slug> with actual values.

Auto-Advance

After the commit is complete, directly run /manage-work advance <ID> to advance to the next stage. The advance command will automatically validate Checkpoint #2 at the E→F boundary.

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.