marketing-book-co-author
Pass
Audited by Gen Agent Trust Hub on Oct 30, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill's primary function is to ingest and transform untrusted user data into first-person narrative, which creates a potential vector for indirect instructions.\n
- Ingestion points: The skill explicitly processes voice notes, bullet fragments, interviews, and rough ideas as source material (
SKILL.md).\n - Boundary markers: There are no explicit instructions or delimiters defined to separate user-provided content from the agent's internal logic, making it susceptible to embedded instructions.\n
- Capability inventory: The skill instructions are limited to text generation, though the execution environment via
run.shinvolves a Python runner.\n - Sanitization: No sanitization or validation protocols for the ingested content are present in the provided instructions.\n- [COMMAND_EXECUTION]: The skill includes a shell script
scripts/run.shto facilitate the execution of the agent.\n - Evidence: The script uses
python3to run a local skill wrapper (run-skill.py) located in the repository's root directory. This is a standard entry point for platform-integrated skills and does not involve elevated privileges or remote script downloading.
Audit Metadata