spark-recipe-newsletter-cleanup
Pass
Audited by Gen Agent Trust Hub on Sep 14, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes external data in the form of email content which could potentially contain malicious instructions designed to trick the agent into performing actions like blocking contacts or unsubscribing without user intent. This risk is inherent to the tool's primary function of processing external communications.
- Ingestion points: The skill reads external email content through the
spark thread <id>command. - Boundary markers: The instructions do not define specific delimiters or guardrails to prevent the agent from following instructions embedded within the email text.
- Capability inventory: The skill utilizes tools to modify email account state, including
blockContact,blockDomain, andunsubscribe. - Sanitization: No sanitization or content filtering logic is implemented within the provided command sequence.
Audit Metadata