briefing-note-expert
Pass
Audited by Gen Agent Trust Hub on Feb 16, 2026
Risk Level: LOWPROMPT_INJECTION
Full Analysis
- PROMPT_INJECTION (LOW): The skill is susceptible to Indirect Prompt Injection (Category 8) due to its core function of synthesizing external data into reports.\n- Ingestion points: Untrusted content is ingested from the
input.jsonfile as specified in the README documentation.\n- Boundary markers: No delimiters or instructions to ignore embedded prompts are documented for the input fields likebackgroundoranalysis.\n- Capability inventory: The skill is restricted to generating Markdown documents in a localReports/directory; it lacks network access, arbitrary command execution, or other high-privilege side effects.\n- Sanitization: While structural validation is performed via JSON schema (validators.py), there is no evidence of natural language sanitization for the data processed into the final briefing note.
Audit Metadata