agentleFS
umararshad4umararshad4/honey-stack/migrate/CLAUDE.md

honey-stack / migrate

CLAUDE.md4 starsChanged 15 days agoMITView on GitHub

honey-stack (plugin honey) supplies the always-on workflow rules: sizing, the ladder, proof before done, decision policy and the receipt line. This file holds only personal preferences.

We split this CLAUDE.md 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:

    Follow explicit user constraints literally.

    line 6
  2. Distinctive:

    Prefer current stable tooling and project-local conventions over inventing a new stack.

    line 7
  3. Distinctive:

    Write clean, scalable, readable code with small modules and clear ownership.

    line 9
  4. Distinctive:

    Do not run full builds in normal sessions unless the change is broad, risky, or the user asks.

    line 10
  5. Very specific:

    Prefer React, Next.js App Router, TypeScript, Tailwind, and shadcn/ui patterns when the project already uses them.

    line 13
  6. Distinctive:

    Keep server/client boundaries explicit.

    line 14
  7. Distinctive:

    Avoid prop drilling when local composition, context, or query/cache patterns are cleaner.

    line 15
  8. Distinctive:

    Keep components focused. If a component grows past roughly 300 lines, consider extraction.

    line 16
  9. Very specific:

    When the user types /graphify, use the installed graphify skill or instructions before doing anything else.

    line 25

≈ “Common” is an estimate: at least 1 in 5 of the 25,680 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.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
CLAUDE.md

Personal instructions

honey-stack (plugin honey) supplies the always-on workflow rules: sizing, the ladder, proof before done, decision policy and the receipt line. This file holds only personal preferences.

Operating principles

  • Follow explicit user constraints literally. If the user says frontend only, vendor only, no code changes, or preserve the UI, treat it as a hard boundary.
  • Prefer current stable tooling and project-local conventions over inventing a new stack.
  • Be pragmatic: say when an idea is risky or not worth it, and offer a cleaner alternative.
  • Write clean, scalable, readable code with small modules and clear ownership.
  • Do not run full builds in normal sessions unless the change is broad, risky, or the user asks.

Frontend defaults

  • Prefer React, Next.js App Router, TypeScript, Tailwind, and shadcn/ui patterns when the project already uses them.
  • Keep server/client boundaries explicit.
  • Avoid prop drilling when local composition, context, or query/cache patterns are cleaner.
  • Keep components focused. If a component grows past roughly 300 lines, consider extraction.
  • Preserve existing UI patterns unless the user asks for redesign.
  • For UI changes, verify responsive behavior when the affected surface is meaningful.

Response style

Be concise. Default to a few lines. No preamble, no recap, no restating the request. Expand only when asked to explain, or when a real risk needs flagging (one or two sentences, then stop).

graphify

  • graphify (~/.claude/skills/graphify/SKILL.md) - any input to knowledge graph. Trigger: /graphify When the user types /graphify, use the installed graphify skill or instructions before doing anything else. Also use the graphify skill proactively (without being asked) when:
  • graphify-out/graph.json exists in the project and the user asks any question about the codebase, its architecture, or file relationships — answer via graphify query first instead of grepping from scratch.
  • The user asks for a map, overview, or architecture explanation of an unfamiliar codebase — offer to build the graph with /graphify if none exists.

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.

Reviews can't be read right now.

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.