Back to skill

Security audit

Demodesk

Security checks for vulnerabilities and agentic risk

Overview

The Demodesk reads fit the stated purpose, but the required CLI can persistently change other agent skills without the skill explaining that impact.

Review and deliberately install a trusted CLI release before use. Disable or explicitly approve its cross-agent skill synchronization, and connect only the intended Demodesk account. Requests use OOMOL as an intermediary. The sole bundled file was fully inspected; the external installers and exact installed CLI behavior remain unverified.

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
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Output HandlingUnvalidated Output Injection, Cross-Context Output, Unbounded Output
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Findings (4)

Description-Behavior Mismatch

High
Category
MCP Tool Poisoning
Confidence
98% confidence
Finding

The declared purpose concerns accessing Demodesk data, but the supplied code executes an external installer. Although CLI installation could be a supporting implementation detail, this chunk establishes no relationship between that installer and Demodesk functionality. The remote script's contents are not supplied, so its further behavior cannot be assessed.

Content

No source excerpt is available for this finding.

External Script Fetching

High
Category
Supply Chain
Confidence
98% confidence
Finding

The command streams a remote installer directly into Bash, making the current contents served by the installation endpoint executable without prior inspection or authenticated version pinning. HTTPS protects transport but does not prevent a compromised vendor endpoint or release from delivering code that accesses local data or modifies the system; this is the same underlying installation risk identified by SQP-2.

Content

Scanner excerpt · SKILL.md (reported line 62)May include surrounding context.

Fallback setup executes downloaded scripts without warning about remote-code execution or system changes.

Medium
Category
Output Handling
Confidence
94% confidence
Finding

The fallback directs the agent to execute downloaded shell or PowerShell code without an explicit approval step, integrity verification, or disclosure of potential system changes. Restricting this to a missing-CLI error reduces exposure, but a compromised installer could still execute arbitrary code with the invoking user's permissions.

Content

No source excerpt is available for this finding.

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 83)May include surrounding context.

md
- **`OOMOL_INSUFFICIENT_CREDIT`** — OOMOL billing stop. Recharge at `https://console.oomol.com/billing/token-recharge` before retrying.

- **HTTP 402 without `OOMOL_INSUFFICIENT_CREDIT`** — the third-party account itself has no active plan or credit. Report the provider error message as is; do not ask the user to recharge OOMOL.

## Resources