ClawHub
Back to skill

Security audit

sunsama

Security checks across malware telemetry and agentic risk

Overview

This skill transparently connects to Sunsama through Maton and asks for user approval before account-changing actions.

Install only if you are comfortable letting Maton mediate access to your Sunsama account and connected calendar/email data. Prefer OAuth over API keys, confirm every write or deletion, and pin the intended Maton profile and Sunsama connection when more than one account is available.

A.I.G

Vulnerability Patterns
  • Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
  • Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
  • Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
  • Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs

SkillSpector

Vulnerability Patterns
  • 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
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (2)

Description-Behavior Mismatch

Medium
Confidence
93% confidence
Finding
The schema exposes a write capability to create a new 'channel/context' even though the skill metadata describes Sunsama task, calendar, backlog, objectives, time tracking, and email-thread operations, not arbitrary channel creation. This kind of capability drift expands the effective permission surface and can let an agent perform unintended state-changing actions beyond the user's expected scope, especially since the metadata also notes that underlying passthrough access may reach more than the documented surface.

Vague Triggers

Low
Confidence
79% confidence
Finding
The description 'Create a new channel/context for tasks' is broad and does not define what a channel is, what system it affects, or what guardrails apply. Ambiguous write-tool descriptions increase the chance an agent invokes the tool inappropriately or a reviewer underestimates its reach, which is more concerning here because this skill already has write-capable integrations and broader passthrough context.

VirusTotal

59/59 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.