meshchatx
Pass
Audited by Gen Agent Trust Hub on Sep 7, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests external data from MeshChatX documentation and GitHub issues through tools like
search_docsand GitHub issue management functions. This creates a surface for indirect prompt injection if the fetched content contains instructions designed to influence the agent's behavior. However, this is an inherent risk for tools that interact with external content and the skill lacks dangerous autonomous capabilities that would escalate this risk. - Ingestion points: GitHub issue content, documentation pages.
- Boundary markers: Not explicitly defined in the high-level description.
- Capability inventory: Documentation search, code snapshotting (read-only), and GitHub issue creation/updates.
- Sanitization: Not specified in the skill manifest.
Audit Metadata