app-namer
Pass
Audited by Gen Agent Trust Hub on Aug 2, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it incorporates external data from
WebSearchandWebFetchtools, as well as user input fromAskUserQuestion, into the agent's context. - Ingestion points: User input defining the app brief and search results from trademark (USPTO, EUIPO) and domain lookups referenced in
SKILL.mdandvalidation-checklist.md. - Boundary markers: There are no explicit instructions or delimiters used to prevent the agent from following instructions that might be embedded in the fetched web content or user-provided app ideas.
- Capability inventory: The skill utilizes
Write,WebSearch, andWebFetchtools, allowing it to record potentially influenced findings toapp-name-shortlist.json. - Sanitization: No sanitization or validation logic is defined to filter the content of external sources before processing.
Audit Metadata