ClawHub
Back to skill

Security audit

Pipedrive API工具

Security checks across malware telemetry and agentic risk

Overview

This appears to be a real Pipedrive API helper, but it needs review because it can affect live CRM records while giving broad and inconsistent usage guidance.

Review before installing. Use only with a least-privileged or test Pipedrive/Maton credential, require explicit confirmation before any create, update, delete, or webhook operation, and avoid printing or pasting API keys into logs or shared transcripts.

SkillSpector

Vulnerability Patterns
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Findings (4)

Intent-Code Divergence

Medium
Confidence
97% confidence
Finding
The recovery example is malformed (`urllib.request.maton.ai/connections')`) and does not perform the documented verification step, while also contradicting the surrounding API usage guidance. Broken troubleshooting for authentication can cause users to misdiagnose failures, improvise unsafe fixes, or bypass expected validation when handling credentials and live API access.

Intent-Code Divergence

Low
Confidence
88% confidence
Finding
The manifest claims managed OAuth support, but the usage example instructs users to send a raw `MATON_API_KEY` bearer token directly. This inconsistency can confuse the trust boundary and encourage use of broader static credentials where users expect delegated OAuth, increasing the chance of credential misuse or over-privileged access.

Vague Triggers

Medium
Confidence
84% confidence
Finding
The skill is described with very broad productivity and workflow-optimization language, making it likely to be selected for generic requests beyond narrow Pipedrive tasks. In agent ecosystems, overbroad routing can cause the skill to receive unrelated sensitive data or trigger external API actions in contexts where the user did not clearly intend CRM access.

Missing User Warnings

Medium
Confidence
93% confidence
Finding
The FAQ and capability descriptions discuss creating, updating, deleting, and webhook-related operations against Pipedrive without prominent warnings that these affect live remote records. In this context, the skill has `exec`/`write` capability and targets a production CRM, so unclear safety boundaries increase the risk of unintended data modification, deletion, or operational changes.

VirusTotal

63/63 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.