okr-design
Pass
Audited by Gen Agent Trust Hub on Mar 9, 2026
Risk Level: SAFENO_CODEPROMPT_INJECTION
Full Analysis
- [NO_CODE]: The skill is entirely composed of Markdown documentation, templates, and facilitation guides. No scripts, binaries, or configuration files that trigger code execution were found.
- [PROMPT_INJECTION]: The skill configuration allows for the use of
WebFetchandWebSearchtools, which serves as an ingestion point for untrusted external data and presents an indirect prompt injection surface. - Ingestion points: The
allowed-toolsmetadata inSKILL.mdincludesWebFetchandWebSearch. - Boundary markers: The skill does not implement delimiters or instructions to treat external content as data only.
- Capability inventory: Available tools are limited to
Read,Glob,Grep,WebFetch, andWebSearch. No high-risk execution capabilities likeevalorsubprocessare present. - Sanitization: The framework does not define validation or filtering for data retrieved via the web tools.
Audit Metadata