Back to skill

Security audit

酷安社区搜索

Security checks for vulnerabilities and agentic risk

Overview

This Coolapk skill is mostly coherent, but it needs review because it uses an unpinned external Python package and documents raw cookie login plus account-changing actions like replies, likes, and follows.

Install only if you are comfortable running the external coolapk-mcp Python package and letting the agent act through your Coolapk account. Treat the cookie as a password, avoid pasting real cookies into shell history or shared transcripts, and review any reply, like, follow, or unfollow command before it runs.

Vulnerability Patterns
  • Insecure DependenciesIntroduces malicious components through unsafe dependency sources
  • 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
Findings (2)

T08 · Insecure Dependencies

Warning
Location
SKILL.md:14
Finding

Unpinned Third-Party Package Installation

Content
View full analysis

Vulnerability Details

File Location: SKILL.md, lines 14-16
Vulnerability Type: Unpinned and unverified external dependency installation
Risk Level: Medium

Vulnerable Code:

bash
pip install coolapk-mcp

Technical Analysis

The Skill instructs users to install coolapk-mcp directly from the package index without specifying an exact version or verifying an integrity hash. The repository does not include the dependency's implementation, a lock file, checksums, a trusted package-index configuration, or another mechanism that binds installation to a reviewed artifact.

Python package installation may execute package-controlled build or installation logic. Subsequent use of the installed coolapk executable also runs code that is outside the audited project. Consequently, the effective behavior can change when a new package release is published, even if this Skill remains unchanged.

README.md also states that the dependency is automatically installed through pip on first use, but the reviewed project contains no implementation showing how that installation is performed.

Attack Path

  1. A user installs or loads the Skill and follows its prerequisite instructions.
  2. The user runs pip install coolapk-mcp.
  3. Pip resolves the package version available from the configured package index at that time.
  4. If the package account, package index, release pipeline, or resolved distribution has been compromised, pip downloads the attacker-controlled artifact.
  5. Package-controlled installation logic or the subsequently invoked CLI executes with the user's operating-system privileges.
  6. The malicious dependency could access files, environment variables, authentication material passed to the CLI, and network resources available to that user.

This path is conditional on compromise or malicious alteration of the external dependency. The audited files contain no evidence that the current package is mal ...[truncated 450 chars]

Remediation
View remediation

Remediation Suggestions

  • Pin the dependency to an exact, reviewed version rather than installing the latest release.
  • Publish a lock file containing cryptographic hashes and install with hash verification, such as pip install --require-hashes.
  • Document the expected package index and prevent fallback to untrusted indexes.
  • Reference the exact source revision corresponding to the pinned distribution.
  • Review the package's source, build configuration, and transitive dependencies before approving upgrades.
  • Prefer installation in an isolated virtual environment with minimal filesystem and network privileges.
  • Do not perform automatic dependency installation without explicit user confirmation and a clear display of the exact version and source.

T09 · Insecure Skill Coding Practices

Warning
Location
SKILL.md:88
Finding

Authentication Cookie Exposed Through Command-Line Arguments

Content
View full analysis

Vulnerability Details

File Location: SKILL.md, lines 88-90
Vulnerability Type: Plaintext sensitive data in process arguments
Risk Level: Medium

Vulnerable Code:

bash
# Login by supplying a Cookie
coolapk login --cookie "uid=xxx;username=xxx;token=xxx"

Technical Analysis

The documented login procedure places the complete authentication cookie in a command-line argument. Command-line secrets can be retained in shell history and terminal transcripts. Depending on the operating system and process configuration, they may also be observable through process-listing utilities, process metadata, audit systems, diagnostic tooling, or command-execution logs.

The cookie contains account authentication material, including a token. Although the example uses placeholders rather than a real secret, users are instructed to replace those placeholders with live credentials. Quoting the value prevents ordinary shell word splitting but does not protect it from history or process-argument disclosure.

Attack Path

  1. A user obtains a valid Coolapk authentication cookie.
  2. The user runs the documented coolapk login --cookie "..." command.
  3. The shell records the command in its history, or local monitoring and process-inspection facilities capture the argument.
  4. Another local principal, support operator, log collector, or process with access to those records retrieves the cookie.
  5. The exposed token is reused before expiration or revocation.
  6. The attacker performs actions permitted by the compromised Coolapk session.

Impact Assessment

Exploitation can result in unauthorized access to the affected Coolapk account for the lifetime and permission scope of the exposed cookie. Based on the documented CLI features, an attacker with a reusable authenticated session may be able to inspect notifications and perform account actions such as posting replies, liking or unliking content, and following or u ...[truncated 177 chars]

Remediation
View remediation

Remediation Suggestions

  • Replace the --cookie argument with hidden interactive input that does not echo the secret.
  • Alternatively, accept the cookie through standard input or a permission-restricted credential file.
  • If environment-variable input is supported, warn that environment variables may also be visible to child processes and diagnostic tooling.
  • Prevent authentication values from being written to application logs, debug output, telemetry, or error messages.
  • Store persistent credentials only when necessary and protect the credential file with owner-only permissions.
  • Document how users can revoke or rotate an exposed cookie.
  • Advise users to remove any existing commands containing live cookies from shell history and relevant execution logs.
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (5)

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
89% confidence
Finding

The README states that the skill can perform login-gated actions such as liking, replying, and following, and that a Python dependency will be automatically installed via pip on first use, but it does not clearly warn users about account-affecting actions or local system modification. In an agent/CLI context, this is dangerous because users may invoke the skill without realizing it can both change their environment and act through their authenticated account, increasing the risk of unintended actions or supply-chain exposure.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The manifest frames the skill as a search/browsing tool, but the documented commands also support authenticated state-changing actions such as like, reply, follow, and unfollow. This mismatch can mislead an agent or user into granting broader trust than warranted, increasing the chance of unintended account actions or abuse when the skill is invoked in automation contexts.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The skill instructs users to pass a raw session cookie directly on the command line, which is highly sensitive because shell history, process listings, logs, agent transcripts, or telemetry may capture it. Since the same skill also exposes account-modifying operations, compromise of that cookie could allow unauthorized actions on the user's Coolapk account.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
76% confidence
Finding

The natural-language instructions and descriptions in this file are entirely in Chinese, with no indication that users may choose another language or that the skill is intentionally limited to a Chinese-speaking audience. Under the policy provided, forcing a specific language without opt-in can be a language or locale policy issue.

Content

No source excerpt is available for this finding.

Rp1

Low
Category
MCP Rug Pull
Confidence
60% confidence
Finding

pip install without ==version installs the latest release, which could include malicious changes.

Content

No source excerpt is available for this finding.