Back to skill

Security audit

Vibe Remote

Security checks for vulnerabilities and agentic risk

Overview

The skill is mostly coherent, but it documents unsafe credential handling and third-party file uploads without enough user safeguards.

Review this skill before installing. Avoid putting real passwords directly in commands, scripts, or CI logs, and only upload issue media after checking the exact file, removing secrets and metadata, and accepting that it will be sent to a third-party host.

Vulnerability Patterns
  • Insecure Skill Coding PracticesFinds exploitable flaws such as hardcoded secrets or command injection
  • 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 (2)

T09 · Insecure Skill Coding Practices

Warning
Location
SKILL.md:27
Finding

Account Password Exposed Through Command-Line Arguments

Content
View full analysis
--password ``` ### Technical Analysis The documented headless invocation passes the account password directly as a command-line argument. Command-line secrets can be exposed through: - Shell history files - CI/CD job definitions and execution logs - Process inspection utilities - Process accounting and monitoring systems - Terminal session recordings - Diagnostic or crash reports that capture process arguments Although access to process arguments may be restricted by the operating system, the password can remain in shell history or logs after the process exits. This creates unnecessary credential exposure beyond the lifetime of the command. ### Attack Path 1. A user follows the documented headless-mode command and substitutes a real password for the placeholder. 2. The shell records the resulting command in its history, or a CI system records it in job configuration or logs. 3. A local user, administrator, monitoring service, compromised process, or party with access to retained logs reads the exposed command. 4. The attacker extracts the account name and password. 5. The attacker authenticates to the affected service using the stolen credentials. 6. Any resulting access is limited by the compromised account's permissions, but it may include unauthorized use of the remote coding environment if the same credentials are accepted by the service. ### Impact Assessment Successful exploitation discloses the supplied account credentials. The attacker may obtain the privileges assigned to the compromised account, potentially including access to remote coding sessions or associated resources. The scope depends on credential reuse, service-side authorization, operating-system process visibilit ...[truncated 206 chars]
Remediation
View remediation
` example and avoid accepting long-lived secrets as command-line arguments. - Prefer an interactive password prompt that disables terminal echo. - For automation, read the password from a protected file descriptor, operating-system credential store, or dedicated secret-management integration. - If environment variables must be supported, document their residual exposure risks and ensure they are masked in CI logs. - Prefer short-lived, narrowly scoped authentication tokens over reusable account passwords. - Add an explicit warning not to place literal credentials in shell scripts, command history, CI configuration, or logs. - Document credential rotation procedures for users who may already have exposed a password this way. ]]>

T09 · Insecure Skill Coding Practices

Warning
Location
SKILL.md:74
Finding

Unrestricted Upload of Local Files to a Third-Party Hosting Service

Content
View full analysis
Remediation
View remediation
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
Category
Not specified by scanner
Confidence
95% confidence
Finding

The skill documents running headless mode with --account and --password on the command line, which can expose credentials through shell history, process listings, CI logs, and monitoring tools. Because this skill is specifically aimed at CI/scripted usage, the surrounding context makes accidental credential disclosure more likely rather than less.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
92% confidence
Finding

The skill instructs users to upload images and videos to catbox.moe, a third-party file hosting service, without any warning that issue attachments may contain sensitive source code, screenshots, tokens, internal URLs, or proprietary data. In a remote coding/support workflow, users are especially likely to share debugging artifacts, so directing them to external hosting increases the risk of unintentional data disclosure.

Content

No source excerpt is available for this finding.