pipedrive
Pass
Audited by Gen Agent Trust Hub on May 2, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses
curlto perform REST API requests against Pipedrive's infrastructure andjqfor parsing JSON responses. It also performs local file writes to/tmp/request.jsonto prepare payload data. - [EXTERNAL_DOWNLOADS]: The skill makes network calls to the official Pipedrive API (
api.pipedrive.com) and references the vendor's own application domain (app.vm0.ai) for connector configuration. - [PROMPT_INJECTION]: The skill presents an indirect prompt injection surface by retrieving and processing external CRM content. 1. Ingestion points: Data retrieved from the Pipedrive API such as deal titles, notes, and contact info in SKILL.md. 2. Boundary markers: No explicit delimiters or instructions to ignore embedded commands are defined for the incoming data. 3. Capability inventory: The skill can execute shell commands via
curland write files to the temporary directory. 4. Sanitization: There is no evidence of sanitization or filtering of the content received from the API before it is handled by the agent.
Audit Metadata