aos-fatos
Pass
Audited by Gen Agent Trust Hub on Aug 12, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill invokes a local helper script (
scripts/aosfatos_helper.py) to manage the assembly of search URLs and the deduplication of retrieved data. This script uses standard Python libraries, performs validation on input parameters (such as year ranges and allowed channel slugs), and encodes search queries to prevent injection issues. - [EXTERNAL_DOWNLOADS]: The skill fetches data from
aosfatos.orgusing the agent's native web fetching capabilities. This interaction is strictly limited to the official domain of the fact-checking service, and the instructions explicitly forbid the agent from using or inventing links from any other source. - [PROMPT_INJECTION]: The skill acknowledges the risk of indirect prompt injection—where instructions might be hidden in external data—and provides a specific guardrail. It instructs the agent to treat all retrieved content as static data to be summarized rather than instructions to be followed, effectively mitigating potential attacks from manipulated web pages.
Audit Metadata