repository-standards

Pass

Audited by Gen Agent Trust Hub on Nov 4, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes a local binary located at .repo-standards/runtime/node_modules/.bin/repo-standards and performs package installation using npm ci. It also instructs the agent to execute commands provided by the tool's output in the event of a version mismatch.
  • [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to "Follow the printed rules" generated by the repo-standards rules command. This creates a vulnerability surface where a compromised repository could provide malicious instructions that the agent would then execute or follow.
  • Ingestion points: Content from the .repo-standards/ directory and the stdout/stderr of the local repo-standards CLI.
  • Boundary markers: None are specified; the agent is explicitly told to follow the output.
  • Capability inventory: Shell command execution via repo-standards and npm, and the ability to modify the filesystem via the adopt command.
  • Sanitization: No sanitization or validation of the rules is performed before the agent is instructed to follow them.
  • [EXTERNAL_DOWNLOADS]: The skill runs npm ci to restore its runtime. While it uses the --ignore-scripts flag to prevent post-installation script execution, it still initiates network connections to download packages from the npm registry.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 4, 2026, 02:26 AM