Skills
skills/near/agent-skills/near-api-js/Gen Agent Trust Hub

near-api-js

Fail

Audited by Gen Agent Trust Hub on Feb 16, 2026

Risk Level: HIGHPROMPT_INJECTIONNO_CODE
Full Analysis
  • [Indirect Prompt Injection] (HIGH): The documentation enables high-privilege blockchain actions. An agent using these as templates for processing untrusted data is at high risk. 1. Ingestion points: key_management.md, tokens_guide.md, and meta_transactions.md. 2. Boundary markers: Absent. 3. Capability inventory: Private key signing, access key rotation, and NEAR/FT transfers. 4. Sanitization: Absent.
  • [Credentials Unsafe] (LOW): Truncated example keys and signatures are found in key_management.md and nep413.md but are clearly for educational use.
  • [No Code] (INFO): The skill contains only Markdown documentation and no executable scripts.
Recommendations
  • AI detected serious security threats
Audit Metadata
Risk Level
HIGH
Analyzed
Feb 16, 2026, 12:18 AM