ops-as-ad-funnel
Pass
Audited by Gen Agent Trust Hub on Nov 2, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructions include a command to run a local utility script,
tools/atoms_query.py, to retrieve marketing 'atoms' from a JSONL file. This is a standard functional tool provided within the skill's repository for data retrieval. - [INDIRECT_PROMPT_INJECTION]: The skill ingests data from local repository files, which constitutes a potential attack surface for indirect prompt injection if those files contain malicious instructions.
- Ingestion points: The script
tools/atoms_query.pyreads data fromknowledge/04-atoms/growth-layers.approved.jsonlinto the agent's context. - Boundary markers: The instructions do not specify any delimiters or safety warnings for handling the content returned from the query script.
- Capability inventory: The skill utilizes local command execution (
python3) to interact with the knowledge base. - Sanitization: No explicit sanitization or validation of the JSONL content is documented.
Audit Metadata