house-style
Pass
Audited by Gen Agent Trust Hub on Oct 28, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze user-provided drafts, creating a potential attack surface where instructions embedded in a draft could attempt to influence the agent's behavior during the audit or configuration process.
- Ingestion points: Untrusted data enters the agent context through the 'Audit' function in
SKILL.md, which takes a draft as input. - Boundary markers: The skill does not explicitly instruct the agent to use delimiters or specific safety markers to differentiate the draft content from its own instructional logic.
- Capability inventory: The skill has the ability to write to the local file system, specifically recording 'House-style deltas' in
.agents/blog-context.md. - Sanitization: There is no evidence of sanitization or validation of the draft content to prevent the injection of malicious instructions into the configuration file.
Audit Metadata