kalshi

Pass

Audited by Gen Agent Trust Hub on Sep 17, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from an external source that could potentially contain malicious instructions or deceptive content.
  • Ingestion points: API responses from external-api.kalshi.com, including market descriptions, event metadata, and platform announcements fetched via WebFetch.
  • Boundary markers: Absent. The instructions do not specify delimiters or warnings for the agent to ignore embedded instructions in the API data.
  • Capability inventory: Bash, Write, and WebFetch tools are available. The skill enables automated trading via POST requests to order endpoints, including balance management and order execution.
  • Sanitization: The provided instructions and Python signing helper do not include logic for sanitizing or validating the content of external API responses before they are processed by the agent.
  • [COMMAND_EXECUTION]: The skill provides complex Python code and Bash scripts for the agent to execute. Specifically, it includes a recursive RSA-PSS signing implementation that processes the KALSHI_PRIVATE_KEY environment variable using regular expressions and dynamic PEM formatting to authenticate requests.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 17, 2026, 04:28 AM
Security Audit — agent-trust-hub — kalshi