codeguard-ruby
Pass
Audited by Gen Agent Trust Hub on Oct 8, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from Ruby source code and project files, which creates an attack surface for indirect prompt injection.\n
- Ingestion points: The skill reads
.rbfiles andGemfilemanifests from the target repository as part of the linting workflow.\n - Boundary markers: No delimiters or instructions to ignore embedded content are used when reading and processing these files.\n
- Capability inventory: The skill has permissions to read local files, modify files (via
rubocop -A), and execute shell commands (rubocop) within the user's environment.\n - Sanitization: There is no mention of sanitizing or escaping the content of files or the paths of files before they are utilized in shell execution or as context for the AI.
Audit Metadata