agentleFS
nelantonenelantone/cognivia/.agents/skills/architecture-audit/SKILL.md

architecture-audit

Skill1 starsChanged 3 months agoMITView on GitHub

Perform a read-only architecture inventory and dependency analysis for frontend, backend, security, RAG, repository, or deployment concerns. Use before cross-cutting implementation, migrations, or architecture decisions that need ownership, coupling, risk, and phased recommendations. Do not use to edit code, execute a refactor, review commits, or produce a task handoff.

15 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use prototype, review and 1 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.
  • GoogleSkill

    prototype

    Build a throwaway prototype to answer a design question. Use when the user wants to sanity-check whether a state model or logic feels right, or explore what a UI should look like.

    Also in 3 projects that use this

  • SentrySkill

    review

    Fixture review skill.

    Also in 3 projects that use this

  • GoogleSkill

    tdd

    Test-driven development. Use when the user wants to build features or fix bugs test-first, mentions "red-green-refactor", or wants integration tests.

    Also in 3 projects that use this

The file

We split this skill into the rules it gives an agent

9 rules. 0 are common, 9 are distinctive, and 0 are very specific to this project.

  • 0

    Common

    Most of the 10 edited versions we read keep these.

  • 9

    Distinctive

    This author's own choices. Worth a look.

  • 0

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Use for a read-only architecture audit of one or more named surfaces: frontend, backend/domain logic, security, RAG, repository structure, …

    line 16
  2. Distinctive:

    Do not use for ordinary file discovery, implementation, refactor execution, documentation synchronization, commit review, or handoff …

    line 21
  3. Distinctive:

    Do not modify, stage, commit, run migrations, install dependencies, or call paid providers.

    line 55
  4. Distinctive:

    Do not inspect or print credential values.

    line 57
  5. Distinctive:

    Do not claim runtime behavior from static inspection alone.

    line 58
  6. Distinctive:

    Do not present a proposed future architecture as current.

    line 59
  7. Distinctive:

    Keep recommendations within the audited surface and explicit non-goals.

    line 60
  8. Distinctive:

    Verify cited paths and separate code evidence, test evidence, documentation, runtime evidence, and inference.

    line 65
  9. Distinctive:

    Check that every proposed phase protects named behavior and has validation and rollback criteria.

    line 67

“Common” is measured: kept by at least 6 in 10 of the 10 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

Architecture Audit

Purpose

Map the current architecture from repository evidence before implementation. Explain ownership, dependencies, coupling, boundaries, and risks without editing the repository.

Trigger

Use for a read-only architecture audit of one or more named surfaces: frontend, backend/domain logic, security, RAG, repository structure, or deployment.

Do not trigger

Do not use for ordinary file discovery, implementation, refactor execution, documentation synchronization, commit review, or handoff generation.

Inputs

  • Audit question, included surfaces, and explicit exclusions.
  • Repository instructions, branch state, and relevant source and tests.
  • Known runtime evidence, documentation, and prior audits.
  • Constraints for compatibility, behavior, security, and deployment.

Workflow

  1. Establish the audit boundary and remain read-only.
  2. Inventory relevant entry points, modules, configuration, tests, assets, and operational files.
  3. Trace imports, calls, data flow, shared state, provider edges, persistence, and deployment dependencies relevant to the question.
  4. Assign each responsibility to its current owner and distinguish intended boundaries from observed coupling.
  5. Record behavior and architecture facts with file evidence; label inference and unknowns explicitly.
  6. Identify concentration points, cycles, hidden state, unsafe trust boundaries, test gaps, and migration constraints.
  7. Recommend ordered, behavior-preserving phases with prerequisites, validation, risks, and rollback points. Do not implement them.

Outputs

Return the inventory, dependency map, ownership map, coupling analysis, risks, and phased recommendations using the architecture audit format.

Safety

  • Do not modify, stage, commit, run migrations, install dependencies, or call paid providers.
  • Do not inspect or print credential values.
  • Do not claim runtime behavior from static inspection alone.
  • Do not present a proposed future architecture as current.
  • Keep recommendations within the audited surface and explicit non-goals.

Validation

  • Cross-check important dependencies in both directions where practical.
  • Verify cited paths and separate code evidence, test evidence, documentation, runtime evidence, and inference.
  • Check that every proposed phase protects named behavior and has validation and rollback criteria.
  • Report excluded or inaccessible surfaces as unverified.

References

  • AGENTS.md — repository boundaries and safety rules.
  • docs/agents/README.md — tooling ownership.
  • Architecture audit format — output structure and surface checklist.

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.