agentleFS
Sign inStart free
ali-sheibaali-sheiba/development-workflow/skills/writing-plans/SKILL.md

writing-plans

Skill1 starsChanged 21 days agoMITView on GitHub

Convert resolved requirements into one reviewable PLAN.md that the user approves and workers execute unit by unit. Use after brainstorming and before any implementation.

187 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use chinese-code-review, chinese-commit-conventions and 10 more.

Other files people use alongside this one.

Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
  • Leodorareluctant259Skill

    chinese-code-review

    中文代码审查规范——在保持专业严谨的同时，用符合国内团队文化的方式给出有效反馈

    Also in 8 projects that use this

  • Leodorareluctant259Skill

    chinese-commit-conventions

    中文 Git 提交规范 — 适配国内团队的 commit message 规范和 changelog 自动化

    Also in 8 projects that use this

  • ryanportfolioSkill

    addskill

    Create, import, update, or install repo or personal skills: runtime ownership, resources, discovery checks.

    Also in 6 projects that use this

  • ryanportfolioSkill

    apply-firmware

    Bring current folder (plain folder, existing repo, old or partial firmware copy) up to full Harness Firmware in place. Mirror external repo → adopt-repo.

    Also in 6 projects that use this

  • +2 more from Leodorareluctant259
  • +6 more from ryanportfolio

The file

We split this skill into the rules it gives an agent

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

  • 0

    Common

    Most of the 169 edited versions we read keep these.

  • 3

    Distinctive

    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:

    Write the one contract the user approves.

    line 8
  2. Very specific:

    Make sure .agent-work/ is ignored (prefer .git/info/exclude).

    line 19
  3. Very specific:

    Create .agent-work/<YYYY-MM-DD>-<feature>/PLAN.md with Status: Draft.

    line 20
  4. Distinctive:

    Do not split for its own sake: every extra unit costs a review.

    line 24
  5. Distinctive:

    Keep it high level with no code blocks.

    line 31
  6. Very specific:

    Ask the user to review PLAN.md.

    line 37

“Common” is measured: kept by at least 6 in 10 of the 169 edited versions of this file we could read. “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

Writing Plans

Write the one contract the user approves. It must let the user catch mistakes before any code exists, and let a worker build one unit without the conversation history. Use the plan format.

Preconditions: brainstorming resolved the requirements, explorers inspected the code, and no implementation has started. If drafting exposes an open requirement, go back to brainstorming - do not choose for the user.

Draft

  1. Make sure .agent-work/ is ignored (prefer .git/info/exclude).
  2. Create .agent-work/<YYYY-MM-DD>-<feature>/PLAN.md with Status: Draft. The folder name never changes.
  3. Split the work into the fewest independently testable units. Each unit names its files and observable acceptance criteria, and ends in one commit. Do not split for its own sake: every extra unit costs a review.
  4. Map every requirement to at least one unit and one verification row.
  5. Name the exact spec and lint commands, with expected results.
  6. Mark each unit User-facing: yes | no. A user-facing unit names its entry point and journey - land on X, act on Y, see Z - and has an end-to-end verification row.

Keep it high level with no code blocks. State outcomes, not implementations. "Add validation" or "handle errors" is not a unit; name what is validated and what the user sees.

Approve

Ask the user to review PLAN.md. Do not branch or implement. Praise, silence, or a request to explain is not approval. Apply changes and ask again until the user explicitly approves. Then set Status: Approved with the approver's name and hand off to development-workflow.

Amend

A change to requirements, scope, units, interfaces, or acceptance criteria: stop the affected units, record the blocker in PROGRESS.md, edit PLAN.md in place, set Status: Draft, and ask for approval again. Resume from the earliest affected unit. Keep one plan per feature - no parallel versions.

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.

Check your setup, free