agentleFS
taecontroltaecontrol/skills/skills/engineering/coding-standards/SKILL.md

coding-standards

Skill1 starsChanged 2 months agoMITView on GitHub

Create, audit, or update a project's durable review standards. Use explicitly when accepted, project-specific coding rules require human judgment and cannot be enforced reliably by tooling.

128 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use everything-claude-code, everything-claude-code-conventions and 10 more.

Other files people use alongside this one.

Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
  • InfopibeSkill

    everything-claude-code

    Also in 9 projects that use this

  • mryeastcodeSkill

    everything-claude-code-conventions

    Development conventions and patterns for everything-claude-code. JavaScript project with conventional commits.

    Also in 8 projects that use this

  • niloykumarbarmanSkill

    agent-sort

    Build an evidence-backed ECC install plan for a specific repo by sorting skills, commands, rules, hooks, and extras into DAILY vs LIBRARY buckets using parallel repo-aware review passes. Use when ECC should be trimmed to what a project actually needs instead of loading the full bundle.

    Also in 8 projects that use this

  • tpavanipradeepSkill

    nextjs-turbopack

    Next.js 16+ and Turbopack — incremental bundling, FS caching, dev speed, and when to use Turbopack vs webpack.

    Also in 15 projects that use this

  • CodeWithBehnamSkill

    dmux-workflows

    Multi-agent orchestration using dmux (tmux pane manager for AI agents). Patterns for parallel agent workflows across Claude Code, Codex, OpenCode, and other harnesses. Use when running multiple agent sessions in parallel or coordinating multi-agent development workflows.

    Also in 17 projects that use this

  • InfopibeSkill

    investor-outreach

    Draft cold emails, warm intro blurbs, follow-ups, update emails, and investor communications for fundraising. Use when the user wants outreach to angels, VCs, strategic investors, or accelerators and needs concise, personalized, investor-facing messaging.

    Also in 22 projects that use this

  • +1 more from Infopibe

The file

We split this skill into the rules it gives an agent

6 rules. 0 are common, 5 are distinctive, and 1 is very specific to this project.

  • 0

    Common

    Most of the 114 edited versions we read keep these.

  • 5

    Distinctive

    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
  1. Very specific:

    Resolve the source: Use the path named by the user, then an established project standards source, otherwise root CODING_STANDARDS.md.

    line 12
  2. Distinctive:

    Filter each candidate: Keep a rule only when it is accepted, project-specific, durable across changes, concrete enough to judge in a diff, …

    line 14
  3. Distinctive:

    Do not install or configure tooling, edit agent instructions, review an implementation, or repair code unless the user separately …

    line 25
  4. Distinctive:

    Do not import a generic style guide or turn current code frequency into desired direction.

    line 26
  5. Distinctive:

    Keep exceptions narrow. Preserve rationale here only when it is needed to apply the exception and no durable decision source owns it better.

    line 27
  6. Distinctive:

    Do not add per-rule owners, status, timestamps, history, or boilerplate sections.

    line 28

“Common” is measured: kept by at least 6 in 10 of the 114 edited versions of this file we could read. “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
SKILL.md

Coding standards

Maintain one concise project source for rules an independent reviewer must judge. Treat observed code as evidence, not policy, and prefer executable enforcement whenever it is reliable.

Process

  1. Resolve the source. Use the path named by the user, then an established project standards source, otherwise root CODING_STANDARDS.md. Read any existing source before proposing changes. Do not create a second canonical file.
  2. Inspect competing owners. Inspect only enough applicable agent instructions, contribution guidance, scripts, CI, formatter, linter, type, test, and build configuration to find contradictions, duplication, or a stronger owner. Inspect code only to understand the consequence and scope of a candidate rule; prevalence does not make a pattern accepted policy.
  3. Filter each candidate. Keep a rule only when it is accepted, project-specific, durable across changes, concrete enough to judge in a diff, and not reliably enforceable elsewhere. Route other information without modifying its owner:
    • machine-checkable rules belong in executable configuration or CI;
    • navigation, non-obvious working choices, hazards, and authority boundaries needed during implementation belong in applicable agent instructions;
    • requirements, architecture, domain meaning, and durable rationale belong in their maintained sources;
    • task findings and preferences without project authority are not standards.
  4. Propose the smallest delta. Show exact additions, revisions, removals, moves to stronger owners, and unresolved conflicts. Prefer deletion when a rule is stale, duplicated, generic, too vague to review, or now enforced automatically. Ask only about material policy choices that available evidence cannot settle.
  5. Obtain acceptance. Do not establish or change project policy without explicit human acceptance of the proposed delta. A review, implementation, or retrospective finding may recommend a candidate rule but cannot add it as a side effect. The exception is a rule delivered through a pull request: refine or garden may add or revise a general rule that defects found in that work demonstrate, and the pull request presents it for the human's review.
  6. Apply and verify. Preserve unrelated content and write only the accepted delta. Use a simple # Coding standards document with ## Review rules and, only when needed, ## Exceptions. Verify that every retained rule has a clear review consequence, every pointer resolves, no stronger executable owner already enforces it, and the repository diff matches the accepted delta.

Boundaries

  • Do not install or configure tooling, edit agent instructions, review an implementation, or repair code unless the user separately authorizes that work.
  • Do not import a generic style guide or turn current code frequency into desired direction.
  • Keep exceptions narrow. Preserve rationale here only when it is needed to apply the exception and no durable decision source owns it better.
  • Do not add per-rule owners, status, timestamps, history, or boilerplate sections. Git preserves document history; omit empty sections.

Outcomes

  • No standards needed: no accepted rule requires a dedicated review-standard source, so no file is created.
  • No changes needed: the existing source is concise, current, and correctly owned.
  • Agree-ready: one minimal delta is ready for human acceptance; no file has changed.
  • Updated: the accepted delta was written and verified at the reported path.
  • Inconclusive: name the exact authority conflict or missing project decision that prevents a safe proposal.

Completion criteria

The project has at most one canonical review-standard source containing only accepted, durable, project-specific rules that require reviewer judgment, or no such file because no rule earns one. Every mutation matches an explicitly accepted delta, and no executable rule or other maintained knowledge was duplicated.

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.

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
What did you use it for? e.g. Reviewing pull requests in a Rails app

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.