ynab
Pass
Audited by Gen Agent Trust Hub on Sep 23, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill utilizes
npxto execute the@useclawlink/clipackage, enabling the agent to perform authentication, retrieve action lists, and execute financial tasks in YNAB. This is a standard and transparent use of command-line tools for the stated service integration. - [INDIRECT_PROMPT_INJECTION]: The skill processes external data from YNAB actions, creating an indirect prompt injection surface common to API integrations. This is mitigated by explicit instructions to the agent to confirm all write operations with the user.
- Ingestion points: Data retrieved from YNAB via the
actions,describe, andruncommands. - Boundary markers: The instructions specify to "confirm with the user before any write."
- Capability inventory: Execution of YNAB actions through a CLI shell interface.
- Sanitization: Relies on user confirmation for state-changing operations.
- [SAFE]: No malicious patterns, obfuscation, or unauthorized data access were detected. All external resources, including the NPM package and project domains, are official infrastructure belonging to the skill's vendor.
Audit Metadata