agentleFS
Sign inStart free
Starlight-IntelligenceStarlight-Intelligence/starlight-agent-skills/skills/coding/starlight-eng-review/SKILL.md

starlight-eng-review

Skill1 starsChanged 4 months agoMITView on GitHub

Stress-test a technical design across data integrity, trust boundaries, concurrency, failure recovery, compatibility, observability, and resource limits. Use before implementing or landing a change whose failure could corrupt state, expose data, break clients, or create operational toil.

1 other project uses a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

3 rules. About 0 are common, 3 are distinctive, and 0 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈3

    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:

    Check invariants, validation, idempotency, atomicity, and concurrency behavior.

    line 28
  2. Distinctive:

    Check compatibility, migrations, rollback, and version negotiation.

    line 31
  3. Distinctive:

    Every blocking finding cites a concrete boundary or contract.

    line 45

≈ “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

Starlight Engineering Review

Purpose

Find architectural failure modes while the design is still cheap to change.

When it fires

  • A change crosses process, network, repository, runtime, database, or trust boundaries.
  • It changes schemas, migrations, authentication, permissions, money, or irreversible state.
  • Reliability, scale, or backward compatibility is material.

Inputs

  • Architecture and data-flow diagrams or equivalent source evidence.
  • State model, API/schema contracts, dependencies, and deployment topology.
  • Expected load, failure history, privacy classification, and recovery objectives.

Workflow

  1. Map inputs, state, side effects, trust boundaries, and owners.
  2. Check invariants, validation, idempotency, atomicity, and concurrency behavior.
  3. Walk failures: timeout, retry, duplicate, partial success, stale state, quota, crash, and dependency drift.
  4. Review authentication, authorization, secret handling, data minimization, and abuse controls.
  5. Check compatibility, migrations, rollback, and version negotiation.
  6. Define observability and tests that falsify the design’s critical assumptions.
  7. Classify findings by severity and recommend blocking landing on unresolved critical risks. The review is advisory; any merge action requires explicit human approval.

Output contract

Return verdict, architecture summary, invariants, trust boundaries, failure table, compatibility/migration findings, resource analysis, required tests, rollback, and residual risks with owners.

Tools & MCP

Inspect actual source, schemas, dependency versions, CI, and deployment configuration. Prefer primary documentation for current protocols. Never paste secrets into the report.

Quality bar

  • Every blocking finding cites a concrete boundary or contract.
  • Distributed and serverless limitations are not hidden by in-process assumptions.
  • Retries and rollbacks are safe, bounded, and testable.
  • “Scalable” and “secure” are replaced by explicit mechanisms and limits.

Example

Input: “Expose a stateless MCP endpoint.”

Good output: review origin enforcement, content limits, version negotiation, request validation, rate-limit topology, serverless state limits, tool authority, cache behavior, and reproducible protocol fixtures.

Built on SIP — Starlight Intelligence Protocol Vertical: starlight-agent-skills · portable capability layer

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.

Check your setup, free