double-check-code

Pass

Audited by Gen Agent Trust Hub on Mar 4, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill instructs the agent to execute standard shell-based development tools such as npm test, pytest, cargo test, and go test. These commands are used to verify code modifications, which is the stated primary purpose of the skill.\n- [PROMPT_INJECTION]: An indirect prompt injection surface is present because the agent executes tests and build scripts within the local project workspace, which may contain user-controlled or external code.\n
  • Ingestion points: Local source code files, package manifests (e.g., package.json), and user-suggested verification scripts.\n
  • Boundary markers: The skill does not implement specific boundary markers or delimiters for the code it processes.\n
  • Capability inventory: The agent has the capability to execute shell commands and interact with the local filesystem through development toolchains.\n
  • Sanitization: The skill explicitly advises negotiating verification methods with the user and considering the use of sandbox environments when standard commands are not applicable, demonstrating a level of security awareness.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 4, 2026, 06:08 PM