runtime-auditor
Pass
Audited by Gen Agent Trust Hub on Jul 31, 2026
Risk Level: SAFENO_CODEPROMPT_INJECTION
Full Analysis
- [NO_CODE]: The skill consists entirely of instructional documentation and metadata. No executable code, shell scripts, or tool definitions are provided.- [PROMPT_INJECTION]: The skill instructions (in
SKILL.mdandmeta/skill.json) direct the agent to process untrusted data sources, including "logs" and "user reports." This ingestion point creates a surface for indirect prompt injection if the audited data contains hidden commands. - Ingestion points: "logs", "traces", "alerts", and "user reports" mentioned in
SKILL.md. - Boundary markers: No delimiters or "ignore instructions" warnings are suggested for processing external data.
- Capability inventory: The skill is instructional only and does not provide new capabilities, though it guides the use of the agent's existing capabilities.
- Sanitization: No sanitization or data validation steps are included in the instructions.
Audit Metadata