kernel
Pass
Audited by Gen Agent Trust Hub on Sep 5, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill acts as an interface for managing and reading from a shared issue repository, creating a potential surface for indirect prompt injection.
- Ingestion points: Data is ingested from the Forge issue store through various
forge issuecommands such aslist,show,search, andrecapwhich read titles, comments, and descriptions. - Boundary markers: The instructions do not define specific delimiters or instructions for the agent to isolate content retrieved from these external issue records.
- Capability inventory: The skill has access to the
Readtool and theBashtool, with the latter restricted to theforge:*command namespace. - Sanitization: There are no explicit instructions for the agent to sanitize or validate the content retrieved from issue metadata before processing it as context for further actions.
Audit Metadata