vibeview
CLAUDE.md4 starsChanged 8 months agoMITView on GitHub
In this repository5 other files it gives its agents.
- claude-code-plugin/CLAUDE.mdCLAUDE.md
- tasks/CLAUDE.mdCLAUDE.md
- add-message-typeSkill
- audit-docsSkill
- auditSkill
The file
We split this CLAUDE.md into the rules it gives an agent
2 rules. About 0 are common, 0 are distinctive, and 2 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈0
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
- Very specific: line 3
Run make check-lite frequently to catch compilation and lint errors early.
- Very specific: line 4
After finishing a task, run make check to verify lint, tests, and build all pass.
≈ “Common” is an estimate: at least 1 in 5 of the 26,980 CLAUDE.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.
Files
Workflow
- Run
make check-litefrequently to catch compilation and lint errors early. - After finishing a task, run
make checkto verify lint, tests, and build all pass. apps/libis a published Go module. CI tags the nextapps/lib/vX.Y.Zwhen it changes onmain, so a change there reaches outside consumers as soon as it lands. A breaking API change must saysdk-bump: minorin its commit message, or it ships as a patch — and versions are immutable once pushed.
Naming
- The project name is always lowercase:
vibeview— never "VibeView" or any other casing, anywhere it appears to users (docs, README, UI, CLI output, plugin/marketplace descriptions, comments). The one exception is where a language requires it, e.g. the Homebrew formula's Ruby classVibeview(Ruby constants must be capitalized).
Coding Principles
- Clarity over cleverness — straightforward solutions, obvious naming, simple control flow.
- Keep it simple (KISS) — start with the simplest design that solves today's problem.
- Design for change — small modules, clear boundaries, minimal coupling.
- Single responsibility — each module/function does one thing well.
- Composition over inheritance — compose behavior from small pieces.
- Explicit interfaces and contracts — make inputs, outputs, and assumptions clear.
- Test behavior, not implementation — focus on critical paths and edge cases.
- Test all new behavior — every new feature or functionality must have a corresponding test.
- One responsibility per file — split by concept, not size.
- Group by feature, not "utils" — related code together, unrelated code apart.
- Backend owns the logic, frontend is a thin display layer — domain logic (state derivation, filtering, classification) lives in Go. The web client consumes server-provided values and handles rendering only.
- Minimal UX for system-like events — low-level session events (hooks, queue operations, last-prompt, etc.) use the shared inline
EventMessagecomponent: compact left-border style with a text preview, click to view raw JSON. Don't introduce custom card/expand UX for these.
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.

