alterlab-hedgefund-monitor
Pass
Audited by Gen Agent Trust Hub on Sep 25, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill queries the official U.S. Office of Financial Research (OFR) API at
data.financialresearch.gov. This is a well-known government domain used for financial stability data. - [COMMAND_EXECUTION]: Provides Python and
curlexamples to interact with the API, process data with Pandas, and create visualizations. These are standard and expected operations for a data-focused skill. - [INDIRECT_PROMPT_INJECTION]: The skill ingests external data from a remote API, creating a surface for indirect prompt injection. The risk is considered low because the source is a verified government domain and the data is retrieved in structured formats.
- Ingestion points: Remote API calls are made in
SKILL.mdandreferences/examples.md. - Boundary markers: None explicitly present in the provided reference scripts.
- Capability inventory: The skill has access to
Bash(python:*)andWebFetch. - Sanitization: Data is parsed as JSON/CSV before being processed into DataFrames.
Audit Metadata