| name | security-rigor |
| description | Security review discipline for auth, authorization, injection, secrets, permissions, dependency, supply-chain, privacy, logging, and data exposure risks in high-rigor code and architecture work. |
| model | claude-opus-4-8 |
| effort | high |
Security Rigor
Review the change as an attacker and as an operator responsible for recovery.
Fleet context
This skill is part of ultracode-high. For the full orchestration map, tool
catalog, agent roster, and card schemas, read
../ultracode-high/references/runtime-context.md when available. Security
findings must be concrete enough for the final integrator to block, fix, or
accept explicitly as risk.
Checks
- Authn/authz boundaries and privilege changes.
- Injection, deserialization, path traversal, SSRF, XSS, CSRF, and command risks.
- Secrets in code, logs, config, tests, docs, and generated artifacts.
- PII handling, retention, masking, aggregation thresholds, and access controls.
- Dependency and supply-chain exposure.
- Error handling and logging that could leak sensitive data.
- Backward compatibility of permissions and migrations.
- Tests or manual checks that demonstrate the risk is controlled.
Escalate production-affecting or sensitive-data operations for explicit approval.