Back to skill

Security audit

Close

Security checks for vulnerabilities and agentic risk

Overview

The Close workflow includes write approvals, but missing-tool setup directs the agent to run unverified third-party installation code without separate approval.

Review and explicitly approve a verified CLI installation, including bundled-skill changes, before allowing setup. Connecting Close gives OOMOL access to CRM requests and credentials. Retain the required write approvals; the remote installer and hosted connector were not verified.

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
94% confidence
Finding

The supplied chunk performs remote script execution for an Oomol CLI installation, not Close data operations. Nothing in the chunk establishes this installation as a supporting dependency for Close. The downloaded script's contents are unavailable, so its additional effects cannot be determined.

Content

No source excerpt is available for this finding.

External Script Fetching

High
Category
Supply Chain
Confidence
98% confidence
Finding

The command pipes a mutable remote script directly into Bash without a pinned version, independently verified integrity check, or review before execution; the PowerShell alternative has the same issue. HTTPS protects transport but does not prevent a compromised distribution endpoint from serving code that can access local files and credentials with the executing user's privileges.

Content

Scanner excerpt · SKILL.md (reported line 70)May include surrounding context.

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

Medium
Category
Output Handling
Confidence
96% confidence
Finding

When the CLI is missing, the skill directs the agent to execute a remote installer without first explaining the local system changes or obtaining explicit installation approval. This condition limits exposure, but a request to operate on Close data does not itself authorize running installation code on the user's machine.

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 91)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