agentleFS
JustineDevsJustineDevs/meta-architect/skills/vet/SKILL.md

vet

Skill5 starsChanged 7 months agoMITView on GitHub

Use when the user wants a security and trust-boundary review of the current design before implementation or release.

9 other projects use a version of this, 1 of them edited.

Adopted by

Most recently taken up by

Used with

Projects with it also use teach.

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

    teach

    Teach the user a new skill or concept, within this workspace.

    Also in 3 projects that use this

The file

We split this skill into the rules it gives an agent

6 rules. 0 are common, 4 are distinctive, and 2 are very specific to this project.

  • 0

    Common

    Most of the 9 edited versions we read keep these.

  • 4

    Distinctive

    This author's own choices. Worth a look.

  • 2

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Use this skill inside Codex to review security posture before the build lane.

    line 8
  2. Very specific:

    Use references/security-playbooks.md when you need the native security playbook set for common trust-boundary reviews.

    line 26
  3. Distinctive:

    Keep security guidance product-owned and lane-aware.

    line 27
  4. Distinctive:

    Call out missing assumptions that affect security posture.

    line 28
  5. Distinctive:

    Test abuse cases such as confused deputy behavior, injection, replay, overbroad access, and failure disclosure.

    line 35
  6. Distinctive:

    Use the SLSA source for build and release supply-chain findings.

    line 44

“Common” is measured: kept by at least 6 in 10 of the 9 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

Vet

Use this skill inside Codex to review security posture before the build lane. $vet remains the sole security gate even as Meta-Architect ships deeper native security playbooks.

Output

Produce:

  • decision, status, evidence, blockers, next_allowed_triggers
  • trust boundaries
  • authn/authz expectations
  • sensitive data paths
  • abuse cases and likely failure modes
  • concrete mitigations
  • accepted risks
  • release blockers vs acceptable risks
  • exact next trigger, usually $vibe

Rules

  • Prioritize material risks over exhaustive but low-value checklists.
  • Use references/security-playbooks.md when you need the native security playbook set for common trust-boundary reviews.
  • Keep security guidance product-owned and lane-aware. Do not introduce a second umbrella or a separate security release gate.
  • Call out missing assumptions that affect security posture.
  • Distinguish between must-fix blockers and documented accepted risk.

Procedure

  1. Enumerate assets, actors, trust boundaries, permissions, and external dependencies.
  2. Trace sensitive inputs from entry to storage, logs, tools, and outbound responses.
  3. Test abuse cases such as confused deputy behavior, injection, replay, overbroad access, and failure disclosure.
  4. Propose least-privilege mitigations with owner, verification, and residual risk.
  5. Route approved remediation to $vibe and keep security gate ownership explicit.

Default evidence dossier

Use docs/reference/security-verification.md as the default OWASP ASVS cross-check for web, API, plugin, and hosted MCP surfaces. Map each material finding to a control and verification artifact. Use the SLSA source for build and release supply-chain findings. Neither source replaces the product threat model, and no standard claim is valid without version-pinned evidence.

Quality bar

  • Findings are tied to a concrete attack path and impact.
  • “Read-only” claims are verified against actual tools, routes, and deployment behavior.
  • Accepted risks include rationale, scope, and a future review trigger.

Example

For a hosted plugin, verify domain ownership, endpoint authentication, tool annotations, skill provenance, secret handling, and whether any route can cause an unapproved side effect.

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.