agentleFS
aradmanamnaoonaradmanamnaoon/vibe-coding-skills/skills/backend/data-consistency-auditor/SKILL.md

data-consistency-auditor

Skill0 starsChanged 19 days agoMITView on GitHub

Review multi-step writes, retries, concurrency, and idempotency for lost updates, duplicates, partial state, and race conditions. Use for payments, jobs, counters, workflows, or shared records.

The file

We split this skill into the rules it gives an agent

7 rules. About 0 are common, 7 are distinctive, and 0 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.

  • 0

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Read applicable repository instructions before edits.

    line 20
  2. Distinctive:

    Write the invariant that must remain true after success, failure, retry, and concurrent execution.

    line 32
  3. Distinctive:

    Look for read-modify-write races, duplicate message/job handling, check-then-act windows, lost updates, and non-idempotent external calls.

    line 34
  4. Distinctive:

    Create deterministic or repeated concurrency tests for the highest-risk interleavings.

    line 36
  5. Distinctive:

    Verify retries converge on the same intended result.

    line 38
  6. Distinctive:

    Use isolated test data; do not induce races or duplicate effects in production.

    line 46
  7. Distinctive:

    If concurrency cannot be reproduced reliably, add invariant instrumentation/stress testing guidance and report the limitation.

    line 63

≈ “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.

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

Purpose

Preserve invariants under retries and concurrent execution, not only the happy path.

When to activate

A workflow writes multiple records/services or can run concurrently/retry.

When not to activate

A pure read-only operation has no state transition.

Preconditions

  • Read applicable repository instructions before edits.
  • Preserve unrelated working-tree changes and establish a baseline when the workflow can modify code.
  • Identify which tools and environments are actually available; do not claim unavailable checks ran.

Required inputs

  • State transitions and invariants
  • Transaction/locking semantics
  • Retry/queue behavior

Workflow

  1. Write the invariant that must remain true after success, failure, retry, and concurrent execution.
  2. Identify the transaction boundary and all side effects that fall outside it.
  3. Look for read-modify-write races, duplicate message/job handling, check-then-act windows, lost updates, and non-idempotent external calls.
  4. Choose an appropriate control: atomic update, transaction, unique/idempotency key, optimistic version, lock, outbox/saga, or compensating action.
  5. Create deterministic or repeated concurrency tests for the highest-risk interleavings.
  6. Inject a failure between steps to verify partial state can recover or is prevented.
  7. Verify retries converge on the same intended result.

Relevant tools

Database/queue docs, integration tests, concurrency/stress harness.

Safety restrictions

Use isolated test data; do not induce races or duplicate effects in production.

Validation

  • The invariant holds after success, injected failure, duplicate request, and tested concurrency interleavings.
  • Controls match the actual storage/message semantics.
  • Residual nondeterministic risk is documented.

Expected output

  • Invariant and transition model
  • Race/partial-state findings
  • Chosen consistency control
  • Failure/concurrency test evidence

Failure handling

If concurrency cannot be reproduced reliably, add invariant instrumentation/stress testing guidance and report the limitation.

Completion criteria

The skill is complete only when every applicable validation item above has observed evidence or is explicitly marked not executed with the concrete blocker. Do not replace missing evidence with confidence language.

Research mapping

This skill was selected from research problems: P034. See research/vibe-coding-problems.md and research/problem-evidence.csv in the Vibe Coding Skills repository.

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.