pipedrive
Pass
Audited by Gen Agent Trust Hub on Sep 4, 2026
Risk Level: SAFECOMMAND_EXECUTIONDATA_EXFILTRATIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill documentation provides numerous Python shell commands using heredocs (
python <<'EOF') to facilitate direct interaction with the Pipedrive API. - [DATA_EXFILTRATION]: The skill performs network operations to
gateway.maton.ai,ctrl.maton.ai, andconnect.maton.ai. It includes theMATON_API_KEY(sourced from the environment) in the Authorization headers for these requests. This behavior is necessary for the skill's stated function of interacting with the Maton gateway. - [INDIRECT_PROMPT_INJECTION]: The skill ingests data from external CRM records via API responses, creating a surface for potential indirect prompt injection if those records contain malicious instructions.
- Ingestion points: API response bodies fetched from
https://gateway.maton.ai/pipedrive/...(documented in SKILL.md). - Boundary markers: There are no specific delimiters or instructional guardrails described to mitigate the execution of instructions embedded in retrieved CRM data.
- Capability inventory: The skill has capabilities for Python shell execution and network access (SKILL.md).
- Sanitization: The instructions do not specify any validation or sanitization routines for the data retrieved from the Pipedrive API.
Audit Metadata