security-best-practice

Pass

Audited by Gen Agent Trust Hub on Apr 26, 2026

Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it processes and follows instructions derived from untrusted project files and source code.\n
  • Ingestion points: The agent scans project files and source code in Step 1 and Step 3 of the SKILL.md file.\n
  • Boundary markers: No explicit delimiters or instructions are provided to the agent to distinguish between its own logic and potential instructions embedded within the user's audited code.\n
  • Capability inventory: The skill is authorized to modify the filesystem and execute shell commands to verify changes in Step 5.\n
  • Sanitization: Input data from the user's project is not sanitized or filtered before being processed by the LLM.\n- [COMMAND_EXECUTION]: The skill invokes system commands to verify code changes.\n
  • Evidence: Step 5 of the instructions directs the agent to execute test runners, linters, and build tools (e.g., npm test, go build) defined within the user's project environment.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 26, 2026, 07:42 PM