agentleFS
jscott3201jscott3201/yet-more-skills/skills/bounded-pr-slice/SKILL.md

bounded-pr-slice

Skill4 starsChanged 2 months agoApache-2.0View on GitHub

Shape a substantial requested change into one reviewable PR outcome with scope, contracts, and acceptance evidence. Not a requirement to write a brief before routine edits.

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 existing focused checks, plus the consumer tests needed by the change.

    line 21

≈ “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

Bounded PR Slice

A slice should be understandable as one outcome, not a count of files or tasks. Use the current repository context and requested goal; the skill itself does not authorize implementation or publication.

Shape the change

Explain what will become possible or correct, what remains out of scope, which contracts matter, and how the result will be demonstrated. Name a prerequisite only when it actually blocks the behavior. Identify source owners and the likely affected components without pretending to know the final implementation before reading them.

Prefer an end-to-end capability over disconnected horizontal work. A protocol correction may include the codec, caller handling, and regression fixture in one PR. A UI workflow may need its API integration and failure state together. Split unrelated outcomes or a migration with independently meaningful stages; do not split a coherent fix into administrative fragments.

Leave room to implement

State invariants and acceptance behavior rather than pseudocode or mandatory helper names. Let the implementer adjust local design after inspecting the code. Revisit the plan when a public contract, data migration, new service, consequential dependency, or ownership conflict changes the promised outcome—not whenever a routine detail differs.

Use existing focused checks, plus the consumer tests needed by the change. Keep heavy native-platform, fuzz, durability, and performance qualification in its established release lane unless the change specifically needs it now. Do not remove required checks to fit a small PR.

For delegated work, identify the owner of shared contracts and generated files. Independent lanes may proceed under the repository's policy; this brief neither requires nor forbids parallel work on its own.

A few paragraphs or an existing issue are usually enough. Hand over the outcome, boundaries, acceptance evidence, and unresolved decisions. Do not require a separate planning file, ledger, or approval for an already clear authorized change.

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.