Back to skill

Security audit

Weibo CLI

Security checks for vulnerabilities and agentic risk

Overview

The skill is mostly a coherent Weibo lookup helper, but it asks users to run an unpinned third-party npm CLI and optionally provide a login cookie without enough security guidance.

Review before installing. Prefer a pinned local installation of the npm package, avoid global install and ad-hoc npx execution, and do not provide WEIBO_COOKIE unless you are comfortable giving the CLI access to an active Weibo session. Treat that cookie like a password and rotate/revoke it if exposed.

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:14
Finding
Unpinned Third-Party npm Package Installation and Execution## Vulnerability Details **File Location**: `SKILL.md`, lines 14-22 **Vulnerability Type**: Unpinned third-party dependency execution **Risk Level**: Medium ### Vulnerable Code ```bash ## Installation # Global install npm install -g @marvae24/weibo-cli # Or local install (safer) npm install @marvae24/weibo-cli npx @marvae24/weibo-cli hot ``` ### Technical Analysis The skill instructs users or agents to install and execute `@marvae24/weibo-cli` without pinning an exact version or providing a lockfile, integrity hash, vendored source, or other mechanism for verifying the retrieved artifact. The audited project contains only `SKILL.md`; therefore, the dependency's implementation and the documentation's claims concerning API access, cookie processing, and retry behavior cannot be verified from the project. Both installation commands resolve a mutable package version from the npm registry. The `npx` command may retrieve and immediately execute that package. Global installation also increases exposure by placing the executable in the user's global command environment. The skill additionally states at line 27 that the package can consume the optional `WEIBO_COOKIE` environment variable. This is not evidence that the current package is malicious, but it means that malicious code introduced through a compromised or changed release could access an authenticated Weibo session cookie when the variable is present. ### Attack Path 1. An attacker compromises the npm package, its publisher account, or a future package release. 2. The attacker publishes malicious code under the same package name and a version selected by npm's unpinned resolution. 3. A user or agent follows the skill instructions and runs `npm install`, `npm install -g`, or `npx`. 4. npm retrieves the mutable package release from the external registry. 5. Package lifecycle scripts or the invoked CLI execute with the privileges of the calling user. 6. The ma ...[truncated 868 chars]
Remediation
## Remediation Suggestions 1. Pin the dependency to an exact, reviewed version rather than relying on the latest registry release. 2. Include a lockfile with npm integrity metadata and use `npm ci` for reproducible installation. 3. Provide a verifiable source-code repository and review the package implementation and lifecycle scripts before recommending execution. 4. Prefer a project-local installation over global installation, and invoke the pinned local binary. 5. Avoid allowing `npx` to download an unverified package at execution time; use an already verified local installation. 6. Execute the CLI in a restricted environment with minimal filesystem and network permissions. 7. Do not expose unrelated environment variables or credentials to the process. 8. If `WEIBO_COOKIE` is necessary, use a dedicated, minimally privileged session, avoid persistent plaintext storage, and rotate the session after suspected exposure. 9. Monitor the dependency for publisher changes, unexpected releases, ownership transfers, and reported supply-chain incidents.
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (4)

Rp1

Medium
Category
MCP Rug Pull
Confidence
91% confidence
Finding
The documentation instructs users to run `npx @marvae24/weibo-cli` without pinning a specific version, which causes execution of whatever package version is current at install time. This creates a supply-chain risk: if the package is compromised or a breaking/malicious version is published, users may execute attacker-controlled code immediately.

Missing User Warnings

Medium
Confidence
96% confidence
Finding
The documentation tells users they can pass a login cookie to the CLI but does not warn that a cookie is effectively account-authentication material. Users may expose an active session token to local logs, shell history, subprocesses, or malicious package code, enabling account takeover or privacy loss if the token is stolen.

Natural-Language Policy Violations

Low
Confidence
73% confidence
Finding
The description frames the skill entirely around Weibo and Chinese content, including Chinese examples and a Chinese social-media platform, without indicating whether users can choose another language/locale. Because the instructions are scoped to a specific locale/domain and do not mention opt-in or alternatives, this may conflict with a language/locale choice policy.

Context-Inappropriate Capability

Low
Confidence
85% confidence
Finding
The skill is presented as a no-auth public-data tool, but it also encourages supplying a personal `WEIBO_COOKIE` for higher rate limits. That widens the trust boundary by inviting users to provide sensitive session material to a third-party CLI, increasing the chance of credential misuse, account compromise, or unintended collection of authenticated data.

Static analysis

No suspicious patterns detected.