inflacion-argentina-ipc
Pass
Audited by Gen Agent Trust Hub on Feb 19, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- EXTERNAL_DOWNLOADS (LOW): The skill makes network requests to
anduin.ferminrp.workers.dev, which is not a trusted external source. - COMMAND_EXECUTION (LOW): The workflow relies on executing
curlandjqin a shell environment to process remote API responses. - PROMPT_INJECTION (LOW): Vulnerable to Indirect Prompt Injection (Category 8) as it ingests untrusted data without boundary markers or sanitization.
- Ingestion points: JSON data from the
/api/v1/indices/inflacion-mensualendpoint. - Boundary markers: Absent. The agent is not instructed to ignore instructions within the fetched content.
- Capability inventory: Shell command execution via
curlandjq. - Sanitization: Absent. Instructions explicitly suggest displaying
error.messageor raw output, which could be exploited to inject instructions.
Audit Metadata