custom-standards
Skill1 starsChanged 5 months agoMITView on GitHub
Project-specific coding standards that OVERRIDE and EXTEND the generic framework and general standards for THIS repository. The code-reviewer, qa-agent, and architect-agent MUST load this skill LAST and treat it as AUTHORITATIVE wherever it conflicts with a framework or general default. Contains only deviations and additions — not a full copy of the framework rules. Ships empty as a template; fill in this project's real rules.
In this repository19 other files it gives its agents.
- CLAUDE.mdCLAUDE.md
- agent-code-to-confirmSkill
- agent-to-confirmSkill
- claude-guideline-creationSkill
- skill-creatorSkill
- youtube-to-guidelineSkill
- youtube-transcribeSkill
- fastapi-standardsSkill
- general-standardsSkill
- go-standardsSkill
- java-standardsSkill
- nextjs-standardsSkill
- nodejs-standardsSkill
- python-standardsSkill
- react-standardsSkill
- reviewing-codebasesSkill
- rust-standardsSkill
- skill-toggleSkill
- typescript-standardsSkill
The file
We split this skill into the rules it gives an agent
4 rules. About 0 are common, 1 is distinctive, and 3 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈1
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: line 24
State management (React): use Redux Toolkit, NOT Zustand (overrides REACT-STATE-4).
- Distinctive: line 34
Every public HTTP endpoint MUST enforce rate limiting.
- Very specific: line 35
Every PR that changes behavior MUST update CHANGELOG.md.
- Very specific: line 36
All feature flags MUST be registered in config/flags.yaml.
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 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
Custom Project Standards
Highest precedence. Every review/QA/architecture agent reads this last, and on any conflict with a framework or general standard, this document wins. Keep it lean: list ONLY where this project diverges from the framework defaults, plus rules that have no framework equivalent. Each override SHOULD name the default it replaces so reviewers can cite it (e.g. "overrides
TS-CONFIG-2"). Agents MUST flag in their reports when a finding is driven by a custom-standards override rather than a framework default.
Overrides (these replace framework/general defaults)
<!-- Examples — delete and replace with this project's real rules:
- **Line length**: 100, overriding the framework default of 88/80.
- **State management (React)**: use Redux Toolkit, NOT Zustand (overrides `REACT-STATE-4`).
- **Package manager**: pnpm only — do not suggest npm/yarn.
- **Indentation**: tabs, not spaces.
-->(none yet — add project overrides here)
Additions (rules with no framework equivalent)
<!-- Examples:
- Every public HTTP endpoint MUST enforce rate limiting.
- Every PR that changes behavior MUST update `CHANGELOG.md`.
- All feature flags MUST be registered in `config/flags.yaml`.
-->(none yet — add project-specific rules here)
Exemptions (paths/rules deliberately exempt, with a reason)
<!-- Examples:
- `legacy/**` is exempt from the type-hint requirement (migration in progress).
- `scripts/oneoff/**` is exempt from the ≥80% coverage rule.
-->(none yet — add exemptions here)
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.