agentleFS
pjlangleypjlangley/ff/.claude/skills/spec-ideate/SKILL.md

spec-ideate

Skill4 starsChanged 52 days agoUnlicenseView on GitHub

Capture a product idea as a requirements spec in specs/. Use when the user wants to ideate, start a new feature / spec, capture requirements, or plan a product ask. Grills the idea via the grilling skill, computes the next zero-padded NNN feature dir, writes requirements.md, and seeds the specs README index.

The file

We split this skill into the rules it gives an agent

1 rule. About 0 are common, 1 is distinctive, and 0 are 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.

  • 0

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Use the bold metadata line under the title (house style), not YAML frontmatter, inside the spec body.

    line 67

≈ “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

spec-ideate

Turn a rough product ask into a requirements.md under specs/NNN_feature_slug/. This is step 1 of the spec-driven workflow: spec-ideate → spec-tasks → spec-build.

When to use

Invoke when the user wants to start a new feature, capture requirements, or "ideate" on a product idea. The input is a product idea for planning purposes — not an implementation request.

Procedure

  1. Grill first. Apply the grilling skill to interrogate the idea: ask one question at a time, wait for each answer, and give a recommended answer per question. Look up facts in the codebase yourself; only put genuine product decisions to the user. Continue until you reach a shared understanding of the product ask.
  2. Compute the next number. List specs/, find the highest existing NNN_* directory, add one, zero-pad to three digits. Numbers are never reused — even if a feature was abandoned, do not fill its slot.
  3. Derive the slug. NNN_feature_slug — snake_case, ASCII only, drop punctuation.
  4. Write the file at specs/NNN_feature_slug/requirements.md using the canonical structure below.
  5. Seed the index. Append a row to the table in specs/README.md with Status: Draft and today's date.
  6. Format. Run deno fmt specs/NNN_feature_slug/requirements.md specs/README.md.
  7. Report back with the new path, and suggest running spec-tasks to break it into deliverable tasks.

Canonical requirements.md structure

Produce a file matching this shape (the leading # is the title, not a section break):

# {Feature title}

**Status:** Draft | **Created:** {YYYY-MM-DD}

## Problem / product ask

{What we're building and why — the user-facing outcome.}

## Users & goals

{Who benefits and what they achieve.}

## Requirements

- {functional requirements — what the feature must do}

## Non-goals / out of scope

- {explicit exclusions to keep scope tight}

## Acceptance criteria

- {observable, testable conditions that define "done"}

## Open questions

- {anything grilling could not resolve — or "none"}

Conventions to honour

  • Zero-padded, never-reused feature NNN (same rule as new-adr).
  • Filenames and directory names are snake_case (project-wide convention).
  • Use the bold metadata line under the title (house style), not YAML frontmatter, inside the spec body.
  • requirements.md is the relatively stable source of truth — later skills read from it rather than rewriting it.

Out of scope

  • Splitting into tasks (that's spec-tasks) or writing any implementation code (that's spec-build).

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.