audit-env-parity
Pass
Audited by Gen Agent Trust Hub on Oct 7, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill functions by ingesting and analyzing untrusted data from the local environment and project source code.
- Ingestion points: The agent is instructed to read
.envfiles,.env.exampletemplates, and application source code to identify environment variable usage (e.g.,process.envorimport.meta.env). - Boundary markers: The instructions do not define specific boundary markers or delimiters to be used when the agent reads and processes the content of these external files.
- Capability inventory: The skill requires file system read capabilities to perform its audit functions across configuration files and the codebase.
- Sanitization: The skill contains explicit instructions to never print secret values and to only output key names in the parity matrix, which provides a level of output sanitization, but it lacks specific input sanitization or validation for the content being analyzed.
Audit Metadata