Back to skill

Security audit

screenata-compliance

Security checks for vulnerabilities and agentic risk

Overview

The skill was rated benign with high confidence: the scanners report a single Markdown component, no AIG findings, and two low-probability findings considered expected for its purpose.

The verdict is benign with high confidence, and Skillspector reports complete coverage of the single component. Before installing, consider that the findings mention a Screenata URL and user-confirmation wording; the scanners judged both expected for the skill’s purpose, but benign does not mean risk-free.

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
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (2)

External Transmission

Medium
Category
Data Exfiltration
Confidence
50% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · SKILL.md (reported line 13)May include surrounding context.

md
## Connect and identify the workspace

This skill supplies instructions, not an MCP connection. If Screenata tools are unavailable, guide the user to add a remote MCP server named `screenata` at `https://api.screenata.com/mcp`, using Streamable HTTP and OAuth. They sign in and select a workspace in the browser. The client must support remote MCP and OAuth discovery. Installation of this skill does not authorize scans, uploads, or workspace changes.

Call `screenata_whoami` and name the workspace before reading its compliance data or making changes. To switch workspaces, reconnect and choose the intended workspace on the consent screen. If sign-in fails, stop workspace operations until it succeeds. Never ask the user to paste tokens or provider credentials into chat.

Autonomous Decision Making

Medium
Category
Excessive Agency
Confidence
80% confidence
Finding

Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.

Content

Scanner excerpt · SKILL.md (reported line 15)May include surrounding context.

md
ction. If Screenata tools are unavailable, guide the user to add a remote MCP server named `screenata` at `https://api.screenata.com/mcp`, using Streamable HTTP and OAuth. They sign in and select a workspace in the browser. The client must support remote MCP and OAuth discovery. Installation of this skill does not authorize scans, uploads, or workspace changes.

Call `screenata_whoami` and name the workspace before reading its compliance data or making changes. To switch workspaces, reconnect and choose the intended workspace on the consent screen. If sign-in fails, stop workspace operations until it succeeds. Never ask the user to paste tokens or provider credentials into chat.

Discover the connected server's current tool schemas; tool names may be prefixed by the client. Use their returned fields and constraints rather than guessing arguments. If a tool reports a missing scope, name it and ask the user to reconnect with that permission. Do not retry an unchanged permission failure.