get-context
Pass
Audited by Gen Agent Trust Hub on Jun 16, 2026
Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [PROMPT_INJECTION]: The skill presents an indirect prompt injection surface through the ingestion of untrusted data.
- Ingestion points: Raw user input via
$ARGUMENTSand content from local dependency manifests (package.json,go.mod,pyproject.toml,Cargo.toml) as described in SKILL.md. - Boundary markers: No delimiters or safety warnings are present to isolate interpolated content from agent instructions.
- Capability inventory: The skill facilitates remote data access via
git cloneand external search tools through a delegated agent. - Sanitization: No input validation or content sanitization is observed.
- [COMMAND_EXECUTION]: The skill enables the execution of commands such as
git cloneand utilizes external research tools (DeepWiki, Context7, Exa) based on user-supplied arguments. This capability can be leveraged to interact with arbitrary remote repositories or services if the delegated agent does not validate the target source.
Audit Metadata