csv-data-summarizer
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- PROMPT_INJECTION (LOW): The SKILL.md file contains forceful instructions designed to override the agent's default behavior, such as 'DO NOT ASK THE USER WHAT THEY WANT', 'DO NOT OFFER OPTIONS', and 'JUST DO IT'. These directives attempt to bypass the agent's standard clarification protocols to force immediate execution.
- Indirect Prompt Injection (LOW): The skill processes untrusted external data (CSV files) and incorporates the results directly into the agent's context. Evidence: 1. Ingestion points: analyze.py reads CSV files via pd.read_csv(). 2. Boundary markers: Absent; there are no delimiters or instructions to ignore embedded content within the processed data. 3. Capability inventory: The skill can write files to the local filesystem (plt.savefig) and perform data processing via pandas. 4. Sanitization: Absent; categorical data and column names are extracted and printed without validation or escaping.
Audit Metadata