Back to skill

Security audit

Browserbase

Security checks for vulnerabilities and agentic risk

Overview

Browserbase changes require approval, but setup can execute unverified installers and introduce broader changes to installed agent skills.

Review and approve CLI installation separately, using a verified release. OOMOL will mediate Browserbase requests. Consider disabling startup skill synchronization with OO_SKILLS_SYNC_DISABLED=1, as documented in the [command reference](https://github.com/oomol-lab/oo-cli/blob/main/docs/commands.md#environment-variables). Keep the skill's approval checks for writes and deletion. Remote installers and hosted runtime behavior 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
  • 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
Findings (4)

Description-Behavior Mismatch

High
Category
MCP Tool Poisoning
Confidence
99% confidence
Finding

The supplied command performs remote script execution for an Oomol installer, not Browserbase data operations. Nothing in the supplied chunk establishes this installer as a supporting Browserbase dependency. The downloaded script's contents and downstream effects are not provided, so the assessment is limited to the visible download-and-execute behavior.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

High
Category
MCP Tool Poisoning
Confidence
84% confidence
Finding

Authentication can be a supporting implementation detail, but the supplied content consists solely of a login command, with no demonstrated connection to Browserbase or implementation of its operations. As supplied, its primary behavior does not match the declared purpose.

Content

No source excerpt is available for this finding.

External Script Fetching

High
Category
Supply Chain
Confidence
99% confidence
Finding

Piping a mutable remote installer directly into Bash executes its contents with the agent process's permissions without an independent integrity check or review. A compromised distribution endpoint could therefore execute arbitrary code and access local data or credentials; HTTPS and the missing-CLI prerequisite narrow exposure but do not prevent this supply-chain attack. The PowerShell installation example has the same underlying weakness.

Content

Scanner excerpt · SKILL.md (reported line 65)May include surrounding context.

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 86)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