agentleFS
intent-hqintent-hq/cloudlands-fe/.agents/skills/pr-description/SKILL.md

pr-description

Skill9 starsChanged 2 days agoApache-2.0View on GitHub

Write short pull request titles and bodies in Simplified Technical English: what changed and why, without text that repeats git or CI.

  • Commits and pushes

97 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use ao-desktop-dev, bug-triage 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.
  • agentlab-inSkill

    ao-desktop-dev

    Launch, restart, or troubleshoot the real AO Electron desktop app from this repository; run a checkout against isolated or real local AO data; combine PR branches for local UI review; and diagnose stale Electron processes, port conflicts, or preload bridge mismatches. Use whenever asked to run, open, show, or visually verify AO frontend changes in the actual desktop app rather than ao preview, dev:web, or mock data.

    Also in 4 projects that use this

  • jMertaSkill

    bug-triage

    Use when the user reports a bug, regression, runtime error, failing build, or broken test and wants reproduction, diagnosis, root-cause analysis, or an authorized fix. Isolates the shared cause and verifies the smallest relevant change.

    Also in 4 projects that use this

  • GitHubSkill

    commit-message

    Generate conventional commit messages - use when creating commits, writing commit messages, or asking for git commit help

    Also in 9 projects that use this

  • saffron-healthSkill

    address-review

    Address PR review comments systematically. Use when responding to code review feedback on a pull request.

    Also in 3 projects that use this

  • NVIDIASkill

    writing-docs

    Use when adding or changing user-facing behavior or documentation in Megatron-Energon, or when creating new .py, .sh or .md files that need license headers.

    Also in 3 projects that use this

  • Sagargupta16Skill

    dependency-audit

    Audit project dependencies for security vulnerabilities, licensing issues, maintenance health, and optimization opportunities. Use it during regular security audits (monthly or quarterly); before a major release when you want to ensure the dependency tree is clean; when evaluating whether to adopt or replace a dependency; after a security advisory affects one of your dependencies; when your install times or bundle sizes have grown noticeably.

    Also in 3 projects that use this

The file

Show its rules and text

We split this skill into the rules it gives an agent

15 rules. 0 are common, 12 are distinctive, and 3 are very specific to this project.

  • 0

    Common

    Most of the 92 edited versions we read keep these.

  • 12

    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. Very specific:

    Use docs only when the diff changes documentation and nothing else.

    line 28
  2. Distinctive:

    Write two parts. No other part is mandatory.

    line 34
  3. Very specific:

    Put the Before image and the After image in the same row.

    line 64
  4. Distinctive:

    Use one row for each state that the PR changes: default, hover, empty, error, long text, narrow width, dark theme.

    line 65
  5. Very specific:

    Use pnpm sandbox:shot when the component has a sandbox scene.

    line 69
  6. Distinctive:

    Do not add a screenshot for a change that has no visible result.

    line 71
  7. Distinctive:

    Write the title and the body in Simplified Technical English (ASD-STE100).

    line 96
  8. Distinctive:

    Write one idea in each sentence.

    line 99
  9. Distinctive:

    Use the active voice. Name the thing that does the action.

    line 100
  10. Distinctive:

    Use the present tense for the new behavior and the past tense for the old.

    line 101
  11. Distinctive:

    Use one word for one thing.

    line 102
  12. Distinctive:

    Use the specific name: the control, the setting, the method, the error.

    line 103
  13. Distinctive:

    Use a verb for an action: "the saga retries the call", not "retry logic was implemented".

    line 104
  14. Distinctive:

    Remove each word that gives no fact: "comprehensive", "robust", "seamless", "leverage", "This PR", "in order to".

    line 106
  15. Distinctive:

    Do not use idioms or metaphors.

    line 108

“Common” is measured: kept by at least 6 in 10 of the 92 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

Pull Request Description

The reviewer reads the diff. The body gives only what the diff cannot show: the change in behavior and the reason for it. Text that repeats git or CI hides those two facts. Do not write it.

The monorepo guide docs/fe/PR_DESCRIPTION_GUIDE.md (intent-hq/intent) has the same rules. Change the two files together.

Title

type(scope): imperative subject

  • CI accepts only these types: feat, fix, chore, docs, refactor, test, ci, perf. The scope is optional.
  • Name the result in 5 to 10 words. Do not end with a period.
  • Use docs only when the diff changes documentation and nothing else.

Example: fix(agent): keep deleted agents out of the sidebar

Body

Write two parts. No other part is mandatory.

  1. What changed. State the new behavior that a user or a developer sees.
  2. Why. State the problem and its cause. If a reviewer can expect a different solution, state why you did not use it.
  • A small change needs one to three sentences and no headings.
  • A large change gets one bullet for each change in behavior.

Add only when it applies

ConditionAdd
The PR fixes a filed issueFixes intent-hq/intent#N
The PR changes UIBefore and after screenshots, in a table
The PR changes what a user sees or doesThe assistant app guide section, or why none applies
The PR claims a performance resultThe measured before and after table, and its command
The PR depends on another PRThe link and the merge order
You ran a check that CI does not runOne line: the check and its result
You did not run a check that is necessaryOne line: the check and the reason
The reviewer must make a decisionOne line: the question and the option that you prefer

Omit a row that does not apply. Do not write "None" or "N/A".

Screenshots

A PR that changes UI shows the change in a table. The reviewer can then compare the two states directly.

  • Attach images with gh pr create --attach ./{file}.png or gh pr edit --attach ./{file}.png; reference ![alt](./{file}.png). Never host images in a gist or on another site.
  • Put the Before image and the After image in the same row.
  • Use one row for each state that the PR changes: default, hover, empty, error, long text, narrow width, dark theme. Name the state in the first column.
  • Capture the two images of a row with the same size, theme, and data.
  • Capture the component, not the full window, unless the PR changes the layout.
  • Use pnpm sandbox:shot when the component has a sandbox scene.
  • New UI has no Before image. Use one column for each theme or width.
  • Do not add a screenshot for a change that has no visible result.
| State   | Before  | After   |
| ------- | ------- | ------- |
| Default | (image) | (image) |
| Error   | (image) | (image) |

Do not write

  • Diff statistics, file counts, or lists of changed, new, or deleted files.
  • Lists of commits, symbols, modules, or components.
  • Results of checks that CI runs: lint, type checks, unit tests, builds.
  • An explanation of how the code works. Put that in the code or in docs/.
  • A list of things that the PR does not change.
  • Risk levels, risk tables, or checklists.
  • Empty sections, placeholders, or a summary of the body.
  • The steps that you took, or the solutions that you tried and removed.
  • Claims that you did not verify.
  • The literal breaking-change footer token, unless the change is a breaking change (see AGENTS.md).

Language: Simplified Technical English

Write the title and the body in Simplified Technical English (ASD-STE100). These rules are mandatory:

  • Write one idea in each sentence. Use a maximum of 20 words.
  • Use the active voice. Name the thing that does the action.
  • Use the present tense for the new behavior and the past tense for the old.
  • Use one word for one thing. Do not change the word to prevent repetition.
  • Use the specific name: the control, the setting, the method, the error.
  • Use a verb for an action: "the saga retries the call", not "retry logic was implemented".
  • Remove each word that gives no fact: "comprehensive", "robust", "seamless", "leverage", "This PR", "in order to".
  • Do not use idioms or metaphors.

Identifiers, commands, and names from the app stay as they are, in code format.

Do not writeWrite
This PR introduces comprehensive improvements to the robustness of the session handling layer.The agent list drops rows that carry pendingDeleteAt.
Retry logic with exponential backoff has been implemented in order to handle transient failures.The saga retries a failed agent.list call three times.
All tests pass.(nothing: CI shows this)

Example

A deleted agent came back in the sidebar after the workspace list refreshed.

The refresh read rows that the daemon had not removed yet. The agent list now
drops rows that carry `pendingDeleteAt`.

Fixes intent-hq/intent#N

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.