executive-brief
Pass
Audited by Gen Agent Trust Hub on Aug 1, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill possesses an indirect prompt injection surface in
SKILL.md(Step 1). It is designed to read and interpolate content fromwork.local.mdinto the agent's context for brief generation. - Ingestion points: Reads
work.local.mdin the current working directory (Step 1). - Boundary markers: None identified. There are no instructions to the agent to treat content within
work.local.mdas untrusted data or to ignore embedded instructions. - Capability inventory: No risky system capabilities (shell execution, network, etc.) were found in the provided files.
- Sanitization: No evidence of escaping or filtering content from
work.local.mdbefore it is used to generate the situation brief.
Audit Metadata