yuki
Pass
Audited by Gen Agent Trust Hub on Sep 30, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent on how to retrieve and process accounting data (invoices, bills, payments) from the Yuki platform, creating a surface for indirect prompt injection if external records contain malicious instructions. 1. Ingestion points: Yuki API entities such as Sales Invoices, Purchase Invoices, and Journal Entries (SKILL.md). 2. Boundary markers: The skill does not define specific delimiters or warnings to ignore embedded instructions in the retrieved data. 3. Capability inventory: The skill utilizes network operations via the Apideck SDK and Proxy API (SKILL.md). 4. Sanitization: There is no mention of sanitization or validation for the incoming external data.
- [COMMAND_EXECUTION]: The skill provides a shell command example using
curlto demonstrate how to use the Apideck Proxy API. This is a standard instructional pattern for this service. - [EXTERNAL_DOWNLOADS]: The skill references the official Node.js package
@apideck/unifyfor integrating with the Apideck unified API. This is a vendor-provided library for the service described.
Audit Metadata