eve-agentic-app-design
Pass
Audited by Gen Agent Trust Hub on Sep 16, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill describes an architecture that processes untrusted data, creating a potential surface for indirect prompt injection.
- Ingestion points: The design includes triggers for external events such as
github.pushandgithub.pull_request, and chat routing from providers likeSlack,Nostr, andWebChat(referenced in SKILL.md). - Boundary markers: The architectural guidance does not define specific delimiters or instructional guardrails for handling untrusted payloads in the provided snippets.
- Capability inventory: The system design empowers agents with
auto_editpermission policies, automated Gitcommitandpushactions, and direct access to environment databases (envdb). - Sanitization: There is no evidence of specific input validation or sanitization requirements for external data before it is processed by the agent personas.
Audit Metadata