agentleFS
ANT1OFFANT1OFF/gram/.agents/skills/mock-review/SKILL.md

mock-review

Skill1 starsChanged yesterdayMITView on GitHub

Take feedback on a mock through to an approved version and hand it to the build. Use when the user comments on, picks between, asks to change, or approves a mock in docs/design/mocks.

2 other projects use a version of this.

Adopted by

Most recently taken up by

  • Other versions2 projects use a version of this.

    Word for word · 0

    None found.

    Edited · 0

    None found.

    Wrote their own version · 2

    1. Immortalqx/my_codex_skills★ 54same name
    2. acprk/crypto-research-agents★ 15same name

    How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.

  • In this repository3 other files it gives its agents.

The file

We split this skill into the rules it gives an agent

9 rules. About 0 are common, 7 are distinctive, and 2 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈7

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 2

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Map the feedback to boards: Use the names on the page ("A Inline", "Empty state").

    line 16
  2. Distinctive:

    When the user picks a direction, the next version develops only that one, at higher fidelity.

    line 17
  3. Very specific:

    Read the current page, Write the next vN.html whole with the Write tool (so it shows in the Files view beside the earlier ones), retake the …

    line 22
  4. Distinctive:

    Ask only when the feedback contradicts the brief or an earlier decision.

    line 33
  5. Distinctive:

    When the user approves ("ship it", "this one", "approved"):

    line 38
  6. Distinctive:

    Check the approved page once at 1280 px and 390 px, light and dark, and fix only a defect (overflow, unreadable text, a state that shows …

    line 40
  7. Very specific:

    Write handoff.md in the mock folder with the Write tool:

    line 43
  8. Distinctive:

    Gaps: every bracketed placeholder still in the page, which the build must fill from real data or raise.

    line 54
  9. Distinctive:

    When the build is up and the user asks to compare, open the approved page and the running app side by side, and list differences by board: …

    line 62

≈ “Common” is an estimate: at least 1 in 5 of the 364,980 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

Mock review

The last step of the mock-first loop: turn feedback into the next version without losing what was already agreed, record each decision, and when the user approves, hand the build a clear target. Mocks are built with mock-build.

A review round

  1. Read the record: the mock's mock.json (which version is current), brief.md and decisions.md, then the current version's page.
  2. Map the feedback to boards. Use the names on the page ("A Inline", "Empty state"). When the user picks a direction, the next version develops only that one, at higher fidelity; the others are dropped, not blended in unless asked.
  3. Change only what was asked. A targeted comment gets a targeted change. Don't restyle, reflow or rename anything else; if something else looks wrong, say so in the reply instead.
  4. New version, never an edit: node .agents/skills/mock-build/scripts/new-mock.mjs <slug> --next, Read the current page, Write the next vN.html whole with the Write tool (so it shows in the Files view beside the earlier ones), retake the changed boards' screenshots in the local shots/, and show it with show_mock (and show_panel preview at vN.html#<board id> for the board that changed). Never hand over a bare path.
  5. Record it: append one line to decisions.md, for example v3: picked B Panel; header quieter; empty state gets the import action (asked 14:20).
  6. Reply with what changed, by board name, and anything you chose not to change and why. A pick between directions is a question with pictures: each option's image and preview, and the mock in mock (mock-build, step 6).

Ask only when the feedback contradicts the brief or an earlier decision; otherwise do the reasonable reading and say which reading you took.

Approval

When the user approves ("ship it", "this one", "approved"):

  1. Check the approved page once at 1280 px and 390 px, light and dark, and fix only a defect (overflow, unreadable text, a state that shows nothing) as one more version, saying so.
  2. node .agents/skills/mock-build/scripts/new-mock.mjs <slug> --approve <n>.
  3. Write handoff.md in the mock folder with the Write tool:
    • Target: the approved file, and the boards and states the build must match.
    • Decisions: the lines from decisions.md that still hold.
    • Parts: each kit part the mock uses, mapped to the real component or the place in the code it becomes. In Gram the components are in apps/web/src/components/ui/: .btn → button, .segmented → button-group, .badge → badge, .callout → alert, .switch → switch, .tabs → tabs, .select → select, .menu → context-menu, title= hints → tooltip, disclosure rows → collapsible; anything with no match (cards, dividers, scrolling boxes) is flagged as new.
    • Deviations allowed: what the build may change without asking (real data lengths, platform controls) and what it may not (structure, hierarchy, the states).
    • Gaps: every bracketed placeholder still in the page, which the build must fill from real data or raise.
  4. Tell the user it's approved, with the approved version's show_mock card, and where the handoff is. The build works from handoff.md and the approved page; a later deviation is recorded in decisions.md, not silently made.

Comparing the build

When the build is up and the user asks to compare, open the approved page and the running app side by side, and list differences by board: matched, deviated with a reason, or missing. Fix what is missing in the build, not in the mock.

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.