Security review

{
  "schemaVersion": 1,
  "id": "security-review",
  "name": "Security review",
  "status": "experimental",
  "purpose": "Review a change for authentication, authorization, data exposure, and injection risk, including risks specific to agent-facing surfaces.",
  "prerequisites": [
    "A concrete diff or branch to review.",
    "Knowledge of which endpoints and data the change touches."
  ],
  "steps": [
    "Check that every new entrypoint authenticates and authorizes at the boundary.",
    "Check that the change grants the least privilege that still works.",
    "Check input validation and output encoding on untrusted data.",
    "Check that secrets come from the environment and never appear in code, config, logs, or errors.",
    "Check injection risk, including content an agent or model may read as instructions.",
    "Check that security-relevant actions leave an audit trail."
  ],
  "capabilities": [
    "read_code",
    "read_docs"
  ],
  "outputs": [
    "Findings describing the vulnerability class and the realistic impact.",
    "A clear statement of which risks were checked and found acceptable."
  ],
  "verification": [
    "Each finding names the entrypoint and the trust boundary it crosses.",
    "No finding, example, or diagnostic reproduces a real credential value."
  ],
  "escalation": [
    "A credential appears to be committed or exposed; stop and notify a human at once.",
    "The finding is exploitable in production; escalate rather than fixing quietly."
  ]
}