kalshi

Pass

Audited by Gen Agent Trust Hub on May 20, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill provides technical documentation and authentication examples for interacting with the Kalshi API. All identified URLs and communication patterns target legitimate Kalshi domains.
  • [CREDENTIALS_UNSAFE]: The skill uses environment variables (KALSHI_ACCESS_KEY, KALSHI_PRIVATE_KEY) to manage authentication secrets. It correctly demonstrates loading these from a .env file, which is a standard security best practice.
  • [DATA_EXFILTRATION]: Sensitive credentials and signatures are only transmitted to the official service provider's endpoints for the purpose of API authentication.
  • [PROMPT_INJECTION]: The skill includes a surface for indirect prompt injection by reading external API content via the WebFetch tool. However, the source (Kalshi) is a regulated financial platform, and the risk of adversarial injection via market descriptions is considered low and manageable by standard model guardrails.
Audit Metadata
Risk Level
SAFE
Analyzed
May 20, 2026, 08:07 AM
Security Audit — agent-trust-hub — kalshi