skills/sebas-aikon-intelligence/antigravity-awesome-skills/product-manager-toolkit/Gen Agent Trust Hub
product-manager-toolkit
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFEPROMPT_INJECTIONNO_CODE
Full Analysis
- PROMPT_INJECTION (LOW): The skill is vulnerable to Indirect Prompt Injection. It describes workflows that involve processing untrusted external data (interview transcripts and feature descriptions) through NLP scripts without evidence of sanitization. Evidence Chain: 1. Ingestion points: scripts/customer_interview_analyzer.py processes interview_transcript.txt and scripts/rice_prioritizer.py processes features.csv. 2. Boundary markers: Absent from provided instructions and usage examples. 3. Capability inventory: The scripts perform NLP extraction, sentiment analysis, and roadmap generation; however, the lack of source code prevents verification of network or filesystem capabilities. 4. Sanitization: No sanitization or escaping of external content is mentioned.
- NO_CODE (SAFE): The skill references executable scripts (scripts/rice_prioritizer.py, scripts/customer_interview_analyzer.py) that are missing from the file package. While this limits functionality, the absence of code itself is not a high-severity security finding unless the instructions prompt the user to download them from an untrusted source.
Audit Metadata