ai-handout-studio
CLAUDE.md1 starsChanged 42 days agoMITView on GitHub
In a Claude cloud session (CLAUDE_CODE_REMOTE=true), the user can't see the screen. Finish every change by opening a pull request as described under "Pull requests" in AGENTS.md, with before/after screenshots when the screen changes. The user asked for this in advance, so open the PR without asking again. In a Claude cloud session, the SessionStart hook (scripts/cloud-session.sh) puts Node 24 from /opt/node24 on PATH, installs dependencies and Chromium; it does nothing locally. The cloud environment's setup script must install Node…
- Commits and pushes
In this repository6 other files it gives its agents.
- AGENTS.mdAGENTS.md
- pull-requestSkill
- studio-setupSkill
- studio-uninstallSkill
- ai-handout-studioSkill
- question-sheetSkill
The file
Files
@AGENTS.md
Claude cloud sessions
In a Claude cloud session (CLAUDE_CODE_REMOTE=true), the user can't see the screen. Finish every change by opening a pull request as described under "Pull requests" in AGENTS.md, with before/after screenshots when the screen changes. The user asked for this in advance, so open the PR without asking again.
In a Claude cloud session, the SessionStart hook (scripts/cloud-session.sh) puts Node 24 from /opt/node24 on PATH, installs dependencies and Chromium; it does nothing locally. The cloud environment's setup script must install Node 24.22.1 to /opt/node24 and pnpm 11.9.0, and its network access must also allow cdn.playwright.dev and playwright.download.prss.microsoft.com.
GitHub Actions
When you run from .github/workflows/claude.yml (GITHUB_ACTIONS=true) and were asked to change code, the same holds: run the checks, and open the pull request yourself with gh pr create --base main as described under "Pull requests" in AGENTS.md, instead of posting a "Create a PR" link. Put Fixes #<issue> in the body's 補足 section. If you were started from a comment on an existing pull request, push to its branch and don't open another one.
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
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.

