Back to skill

Security audit

opencli-browser

Security checks for vulnerabilities and agentic risk

Overview

This skill is a useful browser-control guide, but it normalizes logged-in automation, payment-field inspection, and cached network responses without enough safety boundaries.

Install only if you are comfortable letting an agent drive a real browser session. Avoid using it for passwords, MFA codes, payment data, or high-impact account changes unless you explicitly approve each step, close the browser session when done, and clear the opencli network cache after sensitive work.

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
  • Rogue AgentSelf-Modification, Session Persistence
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (3)

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
92% confidence
Finding

The skill is designed for filling forms and navigating logged-in flows, and even includes a login-form recipe, but it does not provide safety guidance for handling credentials or high-risk account actions. In an agent-facing skill, omission of these guardrails can lead to unsafe automation of authentication flows, mishandling of secrets, or unintended sensitive account changes.

Content

No source excerpt is available for this finding.

Session Persistence

Medium
Category
Rogue Agent
Confidence
82% confidence
Finding

The skill normalizes a persistent live browser automation session and allows it to remain open across commands and until explicit close or timeout. In the context of logged-in browsing, session persistence increases the window in which an agent, another process, or a later task could act within an authenticated context or access residual sensitive state.

Content

Scanner excerpt · SKILL.md (reported line 4)May include surrounding context.

md
---
name: opencli-browser
description: Use when an agent needs to drive a real Chrome window via opencli — inspect a page, fill forms, click through logged-in flows, or extract data ad-hoc. Covers the selector-first target contract, compound form fields, stale-ref handling, network capture, and the agent-native envelopes the CLI returns. Not for writing adapters — see opencli-adapter-author for that.
allowed-tools: Bash(opencli:*), Read, Edit, Write
---

# opencli-browser

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The skill explicitly promotes browser network capture and persistent caching of response bodies in a local cache, including for logged-in sessions, but does not warn that captured API responses may contain tokens, PII, account data, or other sensitive material. Because the cache persists under a user directory and can be re-inspected later, this increases the chance of inadvertent retention and secondary exposure of sensitive session data.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.