agentleFS
refpathrefpath/chatgpt-team/.agents/skills/feature-plan/SKILL.md

feature-plan

Skill0 starsChanged 2 months agoMITView on GitHub

Turn a substantial engineering request into acceptance criteria, risks, interfaces, and a task graph that can be safely delegated.

9 other projects use a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

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

  • 0

    Common

    Most of the 9 edited versions we read keep these.

  • 3

    Distinctive

    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 this skill before substantial implementation or when a request spans several components.

    line 8
  2. Distinctive:

    Read the applicable AGENTS.md files and inspect the repository before proposing work.

    line 12
  3. Distinctive:

    Do not invent cloud services, containers, databases, frameworks, or dependencies merely to make the plan look complete.

    line 39

“Common” is measured: kept by at least 6 in 10 of the 9 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

Feature planning

Use this skill before substantial implementation or when a request spans several components.

Procedure

  1. Read the applicable AGENTS.md files and inspect the repository before proposing work.
  2. Restate the requested outcome as observable acceptance criteria.
  3. Identify affected boundaries: modules, services, browser surfaces, public APIs, protocols, schemas, migrations, security boundaries, and repositories.
  4. Separate facts discovered in the codebase from assumptions that still need confirmation.
  5. Build a dependency graph. Mark tasks as parallel only when they do not require concurrent writes to the same files/branch or an unresolved upstream contract.
  6. Assign the smallest practical role to each task.
  7. Define validation for every task before implementation starts.
  8. Identify which independent review gates apply.

For a changed public or cross-component contract, read contract impact. Include documentation ownership and checks in the task graph. Sequence documentation attributes and runtime edits to the same source file.

Output

For delegated work, include a finite dispatch allowance covering required gates, at most two active children, at most two correction rounds, and run/attempt deadlines. One coordinator owns plan changes; freeze the plan for an evaluation run. See bounded coordination.

Produce a concise plan containing:

  • goal and non-goals;
  • acceptance criteria;
  • affected boundaries;
  • task IDs, owners, dependencies, and expected artifacts;
  • worktree/branch ownership for writers;
  • validation and review gates;
  • unresolved assumptions and material risks.

For a consequential design choice, retain the decision, reason, supporting evidence, and conditions for revisiting it. Use the consuming repository's existing decision format; routine work does not need a new document.

Do not invent cloud services, containers, databases, frameworks, or dependencies merely to make the plan look complete.

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.