Back to skill

Security audit

Agent Browser Clawdbot Bak

Security checks for vulnerabilities and agentic risk

Overview

This is a browser automation helper with disclosed, purpose-aligned capabilities, but saved login state and cookies need careful handling.

Install only if you trust the agent-browser package source. Treat saved auth files, cookies, localStorage, and sessionStorage as credentials: keep them local, do not commit or share them, delete them when no longer needed, and avoid using high-privilege accounts unless the task requires it.

Vulnerability Patterns
  • Insecure DependenciesIntroduces malicious components through unsafe dependency sources
  • 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
Findings (1)

T08 · Insecure Dependencies

Warning
Location
SKILL.md:190
Finding
Unpinned Global Installation of a Third-Party Browser Automation Package## Vulnerability Details **File Location**: `SKILL.md`, lines 190-192 **Vulnerability Type**: Unpinned third-party dependency and unverified component installation **Risk Level**: Medium **Vulnerable Code:** ```bash npm install -g agent-browser agent-browser install # Download Chromium agent-browser install --with-deps # Linux: + system deps ``` ### Technical Analysis The documented installation procedure globally installs the latest available version of the `agent-browser` npm package without an exact version, lockfile, integrity hash, or provenance verification. The newly installed executable is then invoked to download Chromium and, optionally, install system dependencies. Because package resolution is not pinned, the effective code installed can change after the Skill has been reviewed. A compromised maintainer account, malicious package release, registry compromise, or compromised downstream artifact could cause arbitrary package or lifecycle code to execute during installation. Global installation increases the affected scope by exposing the package across the user's environment. The subsequent `install` and `install --with-deps` commands introduce additional supply-chain exposure because they retrieve and install external components without documented checksum or signature validation. ### Attack Path 1. An attacker compromises the npm package, its publisher account, its release pipeline, or a component downloaded by the package. 2. The attacker publishes or substitutes a malicious release while preserving the expected package name. 3. A user follows `SKILL.md` and runs `npm install -g agent-browser` without an exact version constraint. 4. The package manager resolves the attacker-controlled release and executes any applicable installation or lifecycle logic with the installing user's permissions. 5. The user invokes the installed executable to download Chromium or system dependencies ...[truncated 962 chars]
Remediation
## Remediation Suggestions 1. Pin `agent-browser` to a reviewed exact version rather than implicitly installing the latest release. 2. Prefer a project-local dependency recorded in a lockfile over a global installation. 3. Verify npm package provenance, publisher identity, integrity metadata, and release signatures before installation. 4. Review package lifecycle scripts and use installation controls that disable scripts where they are not required. 5. Pin and verify downloaded Chromium artifacts using trusted checksums or signatures. 6. Document the exact external components and versions installed by `install --with-deps`. 7. Perform installation under a dedicated, least-privileged account or isolated environment, and avoid administrative privileges unless explicitly required. 8. Periodically audit pinned versions for known vulnerabilities and update them through a controlled review process.
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Findings (2)

Missing User Warnings

Medium
Confidence
91% confidence
Finding
The skill explicitly recommends saving and reloading browser auth state, which typically includes cookies and other session artifacts that can grant authenticated access. Without any warning about secure storage, scoping, expiration, or prohibition on sharing these files, users or agents may persist sensitive session data insecurely and unintentionally reuse privileged sessions.

Missing User Warnings

Medium
Confidence
87% confidence
Finding
The skill documents commands for reading and modifying cookies and browser storage but omits any notice that these values may contain tokens, PII, or session identifiers. In an agent context, exposing such capabilities without guardrails increases the risk of credential leakage, privacy violations, or unsafe manipulation of authenticated state.

Static analysis

No suspicious patterns detected.