rotation-check
Pass
Audited by Gen Agent Trust Hub on Nov 13, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONNO_CODE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill relies on reading and processing external data from registry and log files to determine audit targets, creating a potential surface for malicious data to influence the agent.
- Ingestion points: The agent is instructed to read
<PIPELINE_ROOT>/<POLICY-DOKUMENTE>,<REGISTRY>, and<LOG>files before selecting a target. - Boundary markers: The prompt template lacks specific delimiters or instructions to treat data as untrusted content, increasing the risk of the agent obeying instructions embedded within these files.
- Capability inventory: The skill directs the agent to perform checks (file reads/audits) and write results to the filesystem (appending to logs and updating the registry).
- Sanitization: No mechanisms for sanitizing or validating the content of the external registry or log files are mentioned.
- [NO_CODE]: The skill consists exclusively of instructional Markdown documentation and prompt templates. No executable scripts, binaries, or automated tools are included in the skill package.
Audit Metadata