agentleFS
openclawopenclaw/openclaw-enterprise/.agents/skills/pr-walkthrough/SKILL.md

pr-walkthrough

Skill454 starsChanged 9 days agoMITView on GitHub

Record a readable Console walkthrough for a pull request. Use when reviewers need to see an interaction, transition, recovery, or keyboard flow.

13 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

4 rules. 0 are common, 3 are distinctive, and 1 is very specific to this project.

  • 0

    Common

    Most of the 12 edited versions we read keep these.

  • 3

    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. Distinctive:

    Start where the change becomes visible.

    line 8
  2. Very specific:

    Call recordWalkthrough from a local TypeScript evidence script.

    line 18
  3. Distinctive:

    Keep these pauses out of product tests.

    line 23
  4. Distinctive:

    Follow the Console visual review instructions to identify the revision, explain what you showed, and share the verified media.

    line 57

“Common” is measured: kept by at least 6 in 10 of the 12 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

Record a pull request walkthrough

Start where the change becomes visible. Show what the reviewer should notice before the action, then leave the result on screen long enough to read it at normal speed. Three to five meaningful actions usually make a 15–30 second video. Use fewer when one interaction explains the change.

Prepare a Playwright page at the viewport you want to show. Finish sign-in and setup before recording, and use disposable data you can share. The video contains the actual page, including visible input values. Label a Storybook scenario as simulated; use the real controller when the claim depends on permissions, persistence, or recovery.

Call recordWalkthrough from a local TypeScript evidence script. Give it a new path in an existing directory outside the checkout. Describe each action and the visible result that proves it worked. The recorder handles the opening pause, transitions, and reading time. It briefly outlines the next control in blue. This is a recording annotation, separate from keyboard focus. Keep these pauses out of product tests.

Replace /path/to/openclaw-enterprise with the absolute path to your checkout so you can run this script from a separate evidence directory.

import { recordWalkthrough } from "/path/to/openclaw-enterprise/scripts/console/record-walkthrough.ts";

const open = page.getByRole("button", { name: "Create Secret" });
await recordWalkthrough(page, {
  outputPath,
  ready: () => open.waitFor({ state: "visible" }),
  steps: [
    {
      target: open,
      action: () => open.click(),
      verify: () =>
        page.getByRole("dialog", { name: "Create Secret" }).waitFor({ state: "visible" }),
    },
  ],
});

action runs once. If a response matters, start waiting for that response inside the action before triggering it. Omit target for an outside event or a keyboard interaction whose focus you need to preserve. verify runs before and after the reading pause. Check the specific visible result, such as a sheet meeting the bottom edge. The recorder waits for ordinary transitions; when the application loads data later, use a check that waits for that data to appear.

Play the saved WebM at normal speed. Its adjacent JSON gives approximate seek times, including when the last result was visible and when the recorder checked it again. Inspect that interval, the opening, and each transition. An incomplete recording can still have changed application state; inspect that state before another take. Follow the Console visual review instructions to identify the revision, explain what you showed, and share the verified media.

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.