agentleFS
MicrosoftKnown publishermicrosoft/cobol-docs-pipeline-sample/.github/skills/requirements-group-writer/SKILL.md

requirements-group-writer

Skill1 starsChanged 55 days agoMITView on GitHub

Author phase-J group ATE (technical-requirements) markdown for ONE group declared in `config/grouping.yaml`. LLM-driven. SCAFFOLD.

The file

We split this skill into the rules it gives an agent

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

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈2

    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:

    Every per-section draft under docs/_shared/_groups/<group_id>/_req-group-sections/<lang>/*.md

    line 47
  2. Distinctive:

    Do not invent requirements for them.

    line 54
  3. Distinctive:

    Write only the languages and paths declared in the bundle.

    line 67

≈ “Common” is an estimate: at least 1 in 5 of the 364,980 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

Status: scaffold — Day-1 contracts only. Mirrors per-file requirements-writer and requirements-finalizer at group scope.

Scope

Owns phase-J authoring at the group level. One invocation per section (per-section scope) or per group (finalizer scope), each covering exactly bundle.languages in declared order at bundle.output_paths. Never add an unrequested English edition or intermediate. Structural parity applies only when multiple languages are requested.

Sibling skills:

  • requirements-writer / requirements-finalizer — per-file ATE (phase F).
  • group-doc-finalizer — group index.md / complete.md (phase I).
  • functional-analysis-group-writer — group AFU (phase K).

Inputs

Per-section scope:

  • docs/_shared/_groups/<group_id>/_req-group-bundles/<section_id>.bundle.json (authoritative) and .bundle.md (LLM-facing projection).

Per-group finalizer scope:

  • docs/_shared/_groups/<group_id>/_req-group-bundles/_finalize.bundle.{json,md}
  • Every per-section draft under docs/_shared/_groups/<group_id>/_req-group-sections/<lang>/*.md

Bundles aggregate, per group:

  • Each eligible member's docs/<lang>/<member>/requirements.md excerpts (phase F).
  • PRC members and members with an upstream _req-bundles/_skipped.json decision are excluded from narrative inputs and recorded in bundle.skipped_members. Do not invent requirements for them. Other missing primary-language member documents are prerequisite errors (exit 4), not implicit skips.
  • Member narrative inputs and bundle excerpts use <lang> = bundle.languages[0]; do not require an English intermediate or member inputs in every output language.
  • The group's docs/<lang>/_groups/<group_id>/complete.md excerpts (phase I).
  • config/grouping.yaml description + member list.
  • The active ATE profile (auto-detected per group).

Outputs

  • docs/_shared/_groups/<group_id>/_req-group-sections/<lang>/<section_id>.md
  • docs/<lang>/_groups/<group_id>/requirements.md

Write only the languages and paths declared in the bundle.

Determinism rules

  • Requirement ids namespace: GRP-<group_id>-REQ-NNN (zero-padded to 3).
  • Citations must point at per-file requirements.md of the member that owns each requirement; never at raw source.
  • Member order in the finalized doc follows config/grouping.yaml.groups[*].members[*] order.

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.