extract-errors

Pass

Audited by Gen Agent Trust Hub on Sep 2, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTIONNO_CODE
Full Analysis
  • Command Execution: The skill includes instructions to run yarn extract-errors. This is a routine development task used to synchronize error codes in React projects. It operates on the assumption that the local environment is correctly configured with the necessary scripts in package.json.
  • Indirect Prompt Injection Surface: The skill processes error messages that might be influenced by external data sources. While this is a common characteristic of text-processing tasks, it is worth noting as a minor area for review.
  • Ingestion points: Error messages and warnings encountered during the React build or extraction process as described in SKILL.md.
  • Boundary markers: The instructions do not explicitly define delimiters to separate error message content from the agent's operational context.
  • Capability inventory: The skill utilizes the agent's ability to execute shell commands via yarn (SKILL.md).
  • Sanitization: No specific filtering or sanitization steps are provided for the error data being analyzed.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 2, 2026, 03:26 PM
Security Audit — agent-trust-hub — extract-errors