tech-survey
Pass
Audited by Gen Agent Trust Hub on Mar 20, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: Indirect Prompt Injection vulnerability surface detected. The skill ingests untrusted data from the web and processes it without adequate security boundaries.\n
- Ingestion points: Phase 1 and Phase 2 utilize the WebFetch tool across multiple sub-agents to retrieve technology-related data from arbitrary external websites.\n
- Boundary markers: The prompt templates for the five sub-agents do not include delimiters or specific instructions to disregard nested commands or instructions within the retrieved content.\n
- Capability inventory: The skill has access to the Write tool (file system operations) and the Bash tool (shell command execution), increasing the risk if an injection occurs.\n
- Sanitization: No sanitization, filtering, or validation of the fetched content is performed before the data is processed by the AI models.
Audit Metadata