harvard-art-museums-etl-pipeline
Pass
Audited by Gen Agent Trust Hub on Oct 1, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill instructions include cloning a specific external repository (
github.com/Manali0711/Harvard-Artifacts-Collection-Data-Engineering-Analytics-App.git) and installing associated dependencies from a requirements file. - [INDIRECT_PROMPT_INJECTION]: The skill fetches and processes data from an external API, which could potentially contain instructions aimed at the agent.
- Ingestion points: The
fetch_artifactsfunction inSKILL.mdretrieves artifact metadata from the Harvard Art Museums API (https://api.harvardartmuseums.org/object). - Boundary markers: There are no explicit delimiters or instructions to the agent to ignore potentially malicious content within the museum records.
- Capability inventory: The skill has the ability to write to a MySQL database and display data via Streamlit as shown in
SKILL.md. - Sanitization: Data is cleaned for basic formatting and length, but no filtering mechanism is implemented to detect or sanitize prompt injection strings.
Audit Metadata