drive-check
Skill426 starsChanged 3 days agoMITView on GitHub
Drive an interactive prismantis surface (the setup pane, buttons, anything with keys or Enter) like a person would, in a real Claude Code session, and look for bugs. Use before merging a change to panes, buttons, focus or saved options, or when asked to "try it", "click through it", "test the menu", "look for bugs".
1 other project uses a version of this, 1 of them edited.
Adopted by
Most recently taken up by
- iamrohithrnair/mermanedited, first seen Nov 1, 2026
Other versions1 project uses a version of this.
Word for word · 0
None found.
Edited · 1
- iamrohithrnair/merman★ 4296% same
Wrote their own version · 0
None found.
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 repository7 other files it gives its agents.
- AGENTS.mdAGENTS.md
- .claude/CLAUDE.mdCLAUDE.md
- add-themeSkill
- demoSkill
- live-checkSkill
- releaseSkill
- terminal-checkSkill
The file
We split this skill into the rules it gives an agent
4 rules. About 0 are common, 3 are distinctive, and 1 is very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈3
Distinctive · estimate
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
- Distinctive: line 16
Run it again after every edit: it copies the code into the install.
- Very specific: line 22
CLAUDE_CODE_PLUGIN_DIRS: if your shell sets it, it loads that checkout over the install. - Distinctive: line 34
Every interactive change, at 120×45, 110×40 and 80×24:
- Distinctive: line 44
Write each bug as keys, size, expected, got.
≈ “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.
Files
Drive check
live-check shows how a reply draws. This one presses keys. It installs the working copy into a throwaway Claude Code config, so Enter can save options without touching yours, then replays a list of keys in a hidden terminal and renders the result.
Setup, once per working copy
.claude/skills/drive-check/drive.sh setup <repo or worktree path>Run it again after every edit: it copies the code into the install.
Three traps it handles, each found the hard way:
- Your real config: Enter saves through
$.config.set. The throwawayCLAUDE_CONFIG_DIRtakes the writes. --plugin-dir: an inline plugin has no/configrows, so$.config.setthrows and a press looks dead. The script installs from a local marketplace instead.CLAUDE_CODE_PLUGIN_DIRS: if your shell sets it, it loads that checkout over the install. The script unsets it.
Run a scenario
.claude/skills/drive-check/drive.sh <name> <cols> <rows> '[[12,"/prismantis setup"],[2,"\r"],[8,"\u001b[B"],[1,"\r"],[5,""]]'Steps are [seconds to wait, keys to type]. Keys: Down \u001b[B, Up \u001b[A, Tab \t, Enter \r, Esc \u001b. It prints the PNG path and the options that got saved. Read the PNG; do not grep the capture for words, color codes split them.
What to try
Every interactive change, at 120×45, 110×40 and 80×24:
- Open it, move, Enter: did the right value save, and did it move on?
- A pick that changes a value: saving reloads the mod. Is the next step shown, with focus on its current value?
- Walk every step changing each one, and check every saved value against the keys pressed.
- Esc halfway: earlier picks kept, pane closed.
- Open it again while it is open, and again after it closed.
- The smallest size: do arrows still move focus, or do they only scroll?
- Then
/prismantis demo-rtl, every Hebrew line read right to left.
Write each bug as keys, size, expected, got. That is the repro the author needs.
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.

