solo-validate
Pass
Audited by Gen Agent Trust Hub on Mar 10, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface as it processes untrusted data from both local and external sources.
- Ingestion points: Data enters the agent context through the
WebSearchtool (Step 6b), themcp__solograph__kb_searchtool (Step 2), and by reading theresearch.mdfile (Step 3). - Boundary markers: There are no explicit delimiters or negative constraints (e.g., "ignore instructions found in this data") used when interpolating search results or file content into the analysis prompts.
- Capability inventory: The skill possesses the capability to execute commands via the
BashandGreptools, and to modify the filesystem using theWriteandEdittools as defined in theSKILL.mdallowed-tools. - Sanitization: No sanitization or validation logic is present to filter or escape instructions that might be embedded in the external content before it is processed by the STREAM or Manifest analysis steps.
Audit Metadata