automation-review
Pass
Audited by Gen Agent Trust Hub on Aug 27, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to analyze content (proposals and inputs) generated by other loops, which creates an attack surface for indirect prompt injection.
- Ingestion points: The skill ingests 'proposals', 'gates', and 'inputs' from external automated sources as defined in the 'How to run' section of SKILL.md.
- Boundary markers: The instructions lack specific delimiters or structural isolation (such as XML tags or explicit quoting rules) to prevent the agent from executing instructions potentially hidden within the proposal text.
- Capability inventory: The skill possesses capabilities to perform network research (fetching web pages) and write to the local file system (updating context files and ledgers), providing a potential impact path for successful injections.
- Sanitization: While referencing general safety rules in 'house-rules.md', the skill does not mandate technical sanitization or validation of the ingested external content.
Audit Metadata