harness-diagnostics

Pass

Audited by Gen Agent Trust Hub on Mar 7, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill includes a shell script (scripts/self-audit.sh) used by the agent to verify the local file structure and integrity of the skill itself. The workflows also suggest executing various CLI tools like rg, npm, and pytest on the project being audited.
  • [PROMPT_INJECTION]: The skill's primary function involves analyzing data from untrusted sources, creating a vulnerability to indirect instructions.
  • Ingestion points: The agent is directed to read files such as AGENTS.md, README.md, and source code from external, potentially untrusted repositories.
  • Boundary markers: The provided instructions do not include delimiters or specific guidance to ignore instructions found within the project files being audited.
  • Capability inventory: The agent has the capability to read any file in the project, scan directories, and execute shell commands for diagnostic purposes.
  • Sanitization: No explicit sanitization or validation of the content of audited files is required before the agent processes the information.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 7, 2026, 07:29 AM