Back to skill

Security audit

Openclaw Quickstart Cn

Security checks for vulnerabilities and agentic risk

Overview

This appears to be a real Chinese OpenClaw quickstart, but it asks users to run unverified internet-delivered installer code and an unpinned global package install.

Install only if you trust the OpenClaw distribution source. Prefer a pinned, versioned package or download-and-verify flow instead of piping a web response into bash. Treat provider API keys as secrets: avoid shared terminals, keep config files out of version control, and rotate any key that may have been exposed.

Vulnerability Patterns
  • Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
  • 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
  • Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
Findings (2)

T03 · Remote Payload Retrieval and Execution

Error
Location
SKILL.md:25
Finding

Unverified Remote Installer Executed Directly by Bash

Content
View full analysis

Vulnerability Details

File Location: SKILL.md, lines 25–27
Vulnerability Type: Remote payload retrieval and execution
Risk Level: High

Vulnerable Code:

bash
# macOS/Linux
curl -fsSL https://get.openclaw.ai | bash

Technical Analysis

The installation instructions pipe the response from https://get.openclaw.ai directly into Bash. The downloaded content is not pinned to an immutable release and is not verified using a cryptographic signature or checksum before execution. Consequently, the effective code executed by users can change after the Skill has been reviewed.

The command also prevents meaningful inspection before execution. Its -s option suppresses ordinary progress and diagnostic output, while the pipe causes Bash to process the response immediately. HTTPS protects the connection in transit under normal conditions, but it does not protect against compromise of the remote domain, hosting infrastructure, deployment pipeline, signing credentials, or legitimate server content.

This behavior exceeds the minimum privileges necessary for a quick-start tutorial. Installation can instead be performed using an immutable, independently verified artifact without allowing mutable network content to flow directly into a command interpreter.

Attack Path

  1. An attacker compromises the remote domain, hosting environment, deployment pipeline, or installer content served from get.openclaw.ai.
  2. The attacker changes the installer response to include malicious shell commands.
  3. A user follows the documented installation command.
  4. curl retrieves the attacker-controlled response and sends it directly to Bash.
  5. Bash executes the payload with the privileges of the user who invoked the command.
  6. The payload may access user-readable files, alter shell configuration, install persistence mechanisms, or obtain credentials available to that user.

Impact Assessment

Successful exploit ...[truncated 593 chars]

Remediation
View remediation

Remediation Suggestions

  1. Remove the direct curl | bash installation pattern.
  2. Link to a versioned, immutable release artifact rather than a mutable installer endpoint.
  3. Publish a cryptographic signature or SHA-256 checksum through an independently controlled channel.
  4. Instruct users to download the artifact without executing it, verify its signature or checksum, and inspect it before installation.
  5. Pin the documentation to a specific audited OpenClaw release.
  6. Run installation with ordinary user privileges unless a narrowly defined operation explicitly requires elevation.
  7. If an installer script remains necessary, provide a transparent sequence such as download, verification, review, and separately confirmed execution.

T08 · Insecure Dependencies

Warning
Location
SKILL.md:29
Finding

Unpinned Global npm Package Installation

Content
View full analysis

Vulnerability Details

File Location: SKILL.md, lines 29–30
Vulnerability Type: Insecure third-party dependency installation
Risk Level: Medium

Vulnerable Code:

bash
# Or use npm
npm install -g openclaw

Technical Analysis

The command installs the package currently resolved under the openclaw name without pinning an exact version or validating package integrity. The installed code can therefore differ from the version considered when this Skill was audited. No lockfile, integrity hash, trusted registry constraint, package provenance check, or publisher verification procedure is supplied.

npm packages may define lifecycle scripts that execute during installation. A compromised maintainer account, malicious publication, registry compromise, or unsafe future release could therefore cause attacker-controlled code to run during installation. The -g option additionally places the package in a global installation scope, which is broader than necessary for a tutorial and may require elevated permissions depending on the local npm configuration.

Attack Path

  1. An attacker compromises the npm package, a maintainer account, or its release process and publishes a malicious version under the expected package name.
  2. A user runs the unpinned command from the documentation.
  3. npm resolves the mutable package name to the malicious release.
  4. npm downloads the package and may execute attacker-controlled lifecycle scripts during installation.
  5. The malicious package gains the privileges of the npm process and can modify resources available to that user.
  6. Because the installation is global, the compromised executable may remain available through the user's command path and affect later OpenClaw invocations.

Impact Assessment

Exploitation could produce arbitrary code execution with the privileges of the user running npm. Potentially exposed resources include user files, environment vari ...[truncated 428 chars]

Remediation
View remediation

Remediation Suggestions

  1. Pin the package to a specific audited version, for example using an exact version rather than a floating package name.
  2. Document the expected npm registry, publisher identity, package provenance, and integrity digest.
  3. Prefer a project-local installation over -g where global scope is not strictly required.
  4. Use a lockfile and integrity metadata where the installation workflow permits it.
  5. Review package lifecycle scripts and use --ignore-scripts when those scripts are not required.
  6. Do not recommend administrator or root execution merely to support a global installation; configure a user-owned installation location instead.
  7. Establish a release-review process before updating the pinned version in the documentation.
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Tool MisuseTool Parameter Abuse, Chaining Abuse, Unsafe Defaults
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (6)

External Script Fetching

High
Category
Supply Chain
Confidence
99% confidence
Finding

The skill explicitly fetches and executes an external script from https://get.openclaw.ai in one step. This creates a high-risk trust boundary violation because users are encouraged to run whatever content the remote endpoint serves at execution time, with no integrity verification or review.

Content

Scanner excerpt · SKILL.md (reported line 27)May include surrounding context.

bash
# macOS/Linux
curl -fsSL https://get.openclaw.ai | bash

# 或使用 npm
npm install -g openclaw

Chaining Abuse

High
Category
Tool Misuse
Confidence
99% confidence
Finding

The | bash construct is dangerous because it chains untrusted network input directly into a shell interpreter. In the context of a beginner quick-start guide, this is more dangerous because inexperienced users are more likely to copy-paste the command without understanding the risks, amplifying the chance of system compromise if the upstream source is malicious or compromised.

Content

Scanner excerpt · SKILL.md (reported line 27)May include surrounding context.

bash
# macOS/Linux
curl -fsSL https://get.openclaw.ai | bash

# 或使用 npm
npm install -g openclaw

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
88% confidence
Finding

The skill content forces a specific language/locale for all users through its title, instructions, and descriptions. Under the stated policy, a language constraint should either be optional for the user or clearly justified as region-specific.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
98% confidence
Finding

The guide instructs users to execute a network-fetched script directly via curl ... | bash, which removes any opportunity to inspect the downloaded code before execution. If the remote host, CDN, DNS, TLS termination, or publishing pipeline is compromised, users could immediately run attacker-controlled code on their systems.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The manifest text repeatedly specifies that the skill is a Chinese-version quickstart guide, including Chinese-only naming and keywords. Under the policy, forcing a specific language or locale without offering a user choice or documenting a justified regional constraint is a natural-language policy violation.

Content

No source excerpt is available for this finding.

Missing User Warnings

Low
Category
Not specified by scanner
Confidence
85% confidence
Finding

The document tells users to configure provider API keys but gives no guidance on secure credential handling, such as avoiding shell history leakage, screenshots, shared terminals, or committing config files to version control. While not an exploit by itself, this omission can lead to credential exposure and subsequent unauthorized API usage or account abuse.

Content

No source excerpt is available for this finding.