NYC

endaoment

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • COMMAND_EXECUTION (SAFE): Shell scripts perform calculations and data formatting using bc, xxd, and jq. All command arguments are properly quoted to ensure safe execution.\n- EXTERNAL_DOWNLOADS (SAFE): Network operations are directed at the official Endaoment API and Base blockchain RPC, which are essential for the skill's purpose.\n- PROMPT_INJECTION (LOW): An indirect prompt injection surface (Category 8) is present because charity metadata is fetched from an external source.\n
  • Ingestion points: Charity names and descriptions are retrieved from api.endaoment.org in scripts/search.sh.\n
  • Boundary markers: None. External strings are displayed raw to the agent.\n
  • Capability inventory: The skill can prepare and prompt for the execution of blockchain transactions via the bankr CLI.\n
  • Sanitization: Input is parsed by jq and truncated, but no filtering for instructions is applied.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 05:15 PM