find-series-opportunities
Pass
Audited by Gen Agent Trust Hub on Apr 16, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill contains instructions to conceal internal processing from the user, such as "read it silently and use it throughout this skill" and "silently rewrite any flagged patterns. Do not mention this step to the user." This reduces transparency in the agent's reasoning process.
- [PROMPT_INJECTION]: The skill is vulnerable to Indirect Prompt Injection because it ingests and processes data from an external library file without sufficient isolation or sanitization.
- Ingestion points: The skill reads content from
~/.devadvokit.md(lines 8 and 20). - Boundary markers: Absent. There are no delimiters or instructions provided to the agent to treat the library content as data rather than instructions.
- Capability inventory: The agent performs pattern matching, thematic clustering, and content strategy generation based on the input file.
- Sanitization: Absent. The skill does not include any validation or filtering mechanisms for the data retrieved from the content library.
Audit Metadata