ooad-discover
Pass
Audited by Gen Agent Trust Hub on Sep 17, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes external project files such as
CONTEXT.mdanddocs/agents/domain.mdto extract domain knowledge and project context. This creates a surface for indirect prompt injection if those files contain adversarial content. - Ingestion points:
SKILL.mdspecifies readingdocs/agents/workflow.md,docs/agents/domain.md, andCONTEXT.mdduring the 'Explore' phase. - Boundary markers: There are no explicit instructions to the agent to treat these files as untrusted or to use delimiters.
- Capability inventory: The skill involves reading and writing markdown files and interacting with GitHub issues via the
ghCLI. - Sanitization: The process does not describe sanitization steps for data read from the repository.
- [EXTERNAL_DOWNLOADS]: The
templates/c4.pumlfile contains a remote include directive for PlantUML definitions. - Evidence:
!include https://raw.githubusercontent.com/plantuml-stdlib/C4-PlantUML/master/C4_Context.puml. - Context: This is a reference to a well-known community-maintained library for C4 diagrams and is documented as a standard external resource.
Audit Metadata