manaflow-aimanaflow-ai/cmux-skills/skills/cmux-artifact/AGENTS.md

cmux-skills / cmux-artifact

AGENTS.md35 starsChanged 5 months agoMITView on GitHub

When asked to make a walkthrough, demo artifact, evidence page, or $cmux-artifact, read the cmux-artifact skill. Create a durable HTML artifact from real evidence, copy supporting files into /cmux-assets/<branch>/..., and open the useful tabs in the caller cmux workspace without stealing focus. Keep the final response short. Include the durable artifact path, whether it was opened in cmux, and the next action the user can take.

We split this AGENTS.md into the rules it gives an agent

5 rules. About 0 are common, 2 are distinctive, and 3 are very specific to this project.

≈0
Common · estimate
Many files like this have a rule of this kind.
≈2
Distinctive · estimate
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

Very specific:
When asked to make a walkthrough, demo artifact, evidence page, or $cmux-artifact, read the cmux-artifact skill.
line 3
Very specific:
Create a durable HTML artifact from real evidence, copy supporting files into /cmux-assets/<branch>/..., and open the useful tabs in the …
line 7
Very specific:
Write index.html with summary, timing table, links to raw evidence, cleanup state, and next actions.
line 13
Distinctive:
Verify the file exists, the artifact is visible in its own pane (not a hidden background tab), and report the durable path.
line 15
Distinctive:
Keep the final response short.
line 19

≈ “Common” is an estimate: at least 1 in 5 of the 36,368 AGENTS.md 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

AGENTS.md

cmux-artifact (Codex agent instructions)

When asked to make a walkthrough, demo artifact, evidence page, or $cmux-artifact, read the cmux-artifact skill.

Your role

Create a durable HTML artifact from real evidence, copy supporting files into /cmux-assets/<branch>/..., and open the useful tabs in the caller cmux workspace without stealing focus.

Checklist

Gather real logs, transcripts, screenshots, videos, timings, or code references.
Copy final evidence out of /tmp into a branch-scoped durable asset directory.
Write index.html with summary, timing table, links to raw evidence, cleanup state, and next actions.
Open index.html as a browser surface (its file:// URL), not cmux open (that makes a file preview, not a live browser). Put it in its own pane side by side with the work: reuse a right helper pane, or in a single-pane workspace create a right split (cmux new-pane --direction right --type browser --url <file-url> --focus false). Never stack it as a background tab behind the active terminal. Open raw evidence (logs, screenshots, video) with cmux open --no-focus after the browser artifact is placed.
Verify the file exists, the artifact is visible in its own pane (not a hidden background tab), and report the durable path.

Output format

Keep the final response short. Include the durable artifact path, whether it was opened in cmux, and the next action the user can take.

Download now (ZIP)

Where files are headingNew AGENTS.md files have more rules about keeping secrets out of the code, and fewer about code style.
Of the AGENTS.md files we read, the 28,693 first seen in the last six months, against the older ones: the kind of rule that gained most share, and the kind that lost most. About files like this in general, not this one, and an estimate, since a kind is a pattern in the words.
In this repository10 other files it gives its agents.
- skills/cmux-cli/AGENTS.mdAGENTS.md
- cmux-artifactSkill
- cmux-browserSkill
- cmux-cliSkill
- cmux-cloudSkill
- cmux-configSkill
- cmux-freestyleSkill
- cmux-refSkill
- cmux-sidebar-builderSkill
- cmux-workspaceSkill

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.

Check your setup, free