sparks
Pass
Audited by Gen Agent Trust Hub on Nov 10, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDATA_EXFILTRATION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests data from the user's workspace (marketing, sales, engineering, etc.) to generate insights.
- Ingestion points: Reviewing work in the conversation and workspace (SKILL.md).
- Boundary markers: Absent. The skill lacks explicit instructions for the model to use delimiters for external workspace data.
- Capability inventory: File access and network transmission to Basecamp via CLI or API tools.
- Sanitization: Robust anonymization rules exist to strip sensitive details (PII, credentials, URLs). A mandatory human-in-the-loop approval step is required before any content is saved.
- [DATA_EXFILTRATION]: The skill transmits captured content to a specific external project management platform.
- Evidence: Target URL is https://app.basecamp.com/5409884/buckets/29731011/message_boards/10495795131.
- Context: Basecamp is a well-known technology service. The operation is the core purpose of the skill and is protected by specific privacy-preserving constraints and user-approval workflows.
Audit Metadata