Back to skill

Security audit

Asyntai AI Chatbot

Security checks for vulnerabilities and agentic risk

Overview

ClawScan rates this skill suspicious with medium confidence because a scanner flagged its Content Security Policy setup advice for allowing inline scripts without warning about the weaker protection.

Review the CSP instructions before installing, and avoid allowing inline scripts unless you understand and accept the reduced protection. Before uploading content or enabling the third-party script, consider what information may be shared with Asyntai and its effect on site visitors. The scan is incomplete because one referenced local path could not be resolved.

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
Vulnerability Patterns
  • Output HandlingUnvalidated Output Injection, Cross-Context Output, Unbounded Output
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (3)

CSP configuration instructions introduce 'unsafe-inline' without warning about weakened security protections.

Medium
Category
Output Handling
Confidence
97% confidence
Finding

Recommending 'unsafe-inline' in script-src can weaken an existing CSP by allowing injected inline JavaScript when nonce/hash protections are absent. Although the external-script alternative reduces the risk, presenting broad inline permission as a routine installation option can remove an important XSS defense; inline style permission also broadens the policy unnecessarily.

Content

No source excerpt is available for this finding.

The setup workflow directs users to upload website content, files, or text to Asyntai without a privacy warning.

Medium
Category
Output Handling
Confidence
96% confidence
Finding

The workflow explicitly instructs users to add content to a third-party service and offers to perform that setup through MCP. It does not warn users to check for confidential or personal information before uploading or explain that the content will be processed outside their own system.

Content

No source excerpt is available for this finding.

Site-wide third-party script installation lacks a privacy and trust warning.

Medium
Category
Output Handling
Confidence
94% confidence
Finding

The instructions direct the agent to add remotely hosted Asyntai JavaScript to the layout shared by every page. The file does not warn that this causes visitors' browsers to contact a third-party service and execute its code in the site's context, which can affect visitor privacy and site integrity.

Content

No source excerpt is available for this finding.