product-lens
Pass
Audited by Gen Agent Trust Hub on Sep 28, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes potentially untrusted external data during its audit modes.
- Ingestion points: Project metadata files such as
README,CLAUDE.md, andpackage.json, as well as recent git commits and product onboarding flows (SKILL.md). - Boundary markers: The instructions lack explicit delimiters or warnings to ignore instructions embedded within the files being analyzed.
- Capability inventory: The skill uses file reading, repository cloning, and shell command execution for product installation (SKILL.md).
- Sanitization: No evidence of input validation or sanitization is present for the data read from external project files.
- [COMMAND_EXECUTION]: Mode 3 ("User Journey Audit") explicitly directs the agent to "Clone/install the product as a new user". This instruction requires the agent to execute shell commands to download and run external software, which is a powerful capability that could be abused if directed at malicious repositories.
Audit Metadata