mesh-recon
Pass
Audited by Gen Agent Trust Hub on Jul 9, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill employs the Bash tool to inspect service mesh configuration files such as PeerAuthentication and VirtualService.
- [EXTERNAL_DOWNLOADS]: The skill is configured with WebFetch and WebSearch capabilities for external data retrieval.
- [PROMPT_INJECTION]: The skill presents a surface for indirect prompt injection (Category 8) by processing external configuration files. 1. Ingestion points: Step 1 reads Istio/Linkerd and PeerAuthentication configs from the environment. 2. Boundary markers: No specific delimiters or safety warnings are defined to separate config content from agent instructions. 3. Capability inventory: The skill has access to Bash, Write, and WebFetch tools which could be leveraged if malicious instructions were present in audited files. 4. Sanitization: No evidence of sanitization or validation of the ingested configuration content is provided.
- [SAFE]: The behavior of the skill is consistent with its stated purpose of infrastructure auditing and the author's identity. No malicious patterns were identified.
Audit Metadata