agentleFS
SoftBlack42SoftBlack42/StarPie/.agents/skills/feature-planning/SKILL.md

feature-planning

Skill1.3k starsChanged 14 days agoMITView on GitHub

Plan bounded StarPie features with architecture placement, compatibility rules, verification gates, worktree scope, and GPT-to-Gemini handoff contracts. Use before multi-file implementation or delegated development.

10 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, 2 are distinctive, and 1 is very specific to this project.

  • 0

    Common

    Most of the 10 edited versions we read keep these.

  • 2

    Distinctive

    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:

    Always read the root AGENTS.md and docs/architecture.md.

    line 10
  2. Distinctive:

    Prefer additive, reversible changes.

    line 25
  3. Distinctive:

    After the implementation request is visibly delivered and work starts, end the Codex turn.

    line 31

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

StarPie Feature Planning

Read selectively

Always read the root AGENTS.md and docs/architecture.md. Then load only the affected domain documents listed in the root routing table.

Produce a bounded plan

Define:

  • user outcome and non-goals;
  • repository, isolated worktree, base ref and existing dirty state;
  • owning layer and why the work belongs there;
  • allowed and forbidden paths;
  • compatibility, latency, memory, security/capability and localization impacts;
  • implementation stages with observable deliverables;
  • automated gates, independent review and human acceptance;
  • completion and stop conditions, meaningful progress checks and minimum useful incomplete outcome; record numeric budgets only when the user explicitly sets them for this task. Do not impose model-call, conversation-turn, delegated-command, repair-cycle or delegation-time quotas by default.

Prefer additive, reversible changes. Reuse existing fact sources, renderers, forms, host services and action queues instead of creating parallel systems.

GPT to Gemini handoff

The handoff must include exact worktree, base commit, model profile, task ID/revision, allowed scope, required commands, evidence format and prohibited actions. Do not authorize commit, push, merge, Tag or Release unless the user separately did so.

After the implementation request is visibly delivered and work starts, end the Codex turn. Do not poll, sleep, monitor logs or spend tokens on routine status. Resume only after the user reports completion, asks for progress, or requests troubleshooting.

Continue authorized delegation, repairs and review as needed to close the task. Do not require renewal solely because a self-imposed call, turn, command, repair-cycle or delegation-time quota was reached. Explicit user budgets, path/permission boundaries and stop conditions still apply; sustained lack of progress calls for a changed approach or a clear report, not empty retries.

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.