spec-conformance-audit
Pass
Audited by Gen Agent Trust Hub on Sep 10, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data including specifications and code diffs.
- Ingestion points: The skill processes 'named specification, issue, plan' and 'relevant diff, current code' as described in SKILL.md.
- Boundary markers: The instructions command the agent to 'Cite the exact contract source and implementation location', providing source attribution, though no explicit prompt delimiters are defined.
- Capability inventory: The skill is explicitly restricted to 'Audit only' and contains instructions forbidding the agent from editing files, applying fixes, or mutating external systems.
- Sanitization: No explicit sanitization or filtering of ingested content is defined in the instructions.
- [SAFE]: No malicious code, persistence mechanisms, exfiltration patterns, or unauthorized tool usage were identified in the skill instructions or configuration files.
Audit Metadata