recursive-boil-down
Pass
Audited by Gen Agent Trust Hub on Sep 27, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process content from external source repositories and configuration files, creating a vulnerability surface where malicious instructions embedded in those files could influence agent behavior.
- Ingestion points: The protocol involves reading the target source tree, manifest files (seed.yaml), and repository inventories as described in SKILL.md.
- Boundary markers: The instructions do not provide explicit delimiters or 'ignore' commands to separate processed data from the agent's core instructions.
- Capability inventory: According to the frontmatter, the skill utilizes reads-filesystem and creates-files.
- Sanitization: There is no evidence of content sanitization or validation before the agent analyzes the ingested source files.
Audit Metadata