Back to skill

Security audit

Minimax Text

Security checks for vulnerabilities and agentic risk

Overview

This skill is a straightforward MiniMax text-generation wrapper that uses a user-provided API key to call the MiniMax API.

Install only if you are comfortable sending prompts, system messages, and conversation history to MiniMax. Use a scoped MiniMax API key, avoid placing secrets in prompts, and prefer an isolated Python environment with pinned dependencies.

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

Note
Location
SKILL.md:24
Finding

Unpinned Third-Party Dependency Installation

Content
View full analysis

Vulnerability Details

File Location: SKILL.md:24
Vulnerability Type: Unpinned runtime dependency
Risk Level: Low

Vulnerable Code Snippet:

bash
pip3 install requests

Technical Analysis

The installation instructions retrieve the latest available release of requests without a pinned version, cryptographic hashes, a lockfile, or an index restriction. This makes installation results dependent on mutable package-registry state and prevents users from verifying that they are installing the exact dependency version reviewed by the Skill publisher.

The instruction references the legitimate requests package rather than an observed typosquat or malicious source. Therefore, this is a supply-chain hardening weakness, not evidence that the Skill intentionally installs malware.

Attack Path

  1. An attacker compromises the package, a relevant package-distribution account, the configured Python package index, or the dependency delivery path.
  2. A user follows the documented pip3 install requests instruction.
  3. pip resolves and downloads whatever release or artifact the mutable source currently serves.
  4. Malicious package installation behavior or package code executes in the user's environment.
  5. The payload gains the privileges of the account running pip or later invoking the Skill.

This path requires an upstream supply-chain or distribution compromise; the audited project does not itself provide a malicious package.

Impact Assessment

Successful exploitation could execute arbitrary Python code with the installing or runtime user's privileges. Depending on that user's access, the attacker could read or modify accessible files, environment variables, and application data, including MINIMAX_API_KEY, and make network requests. No privilege-escalation mechanism or persistence behavior was found in the audited Skill itself.

Remediation
View remediation

Remediation Suggestions

  • Declare dependencies in a reviewed requirements or lock file.

  • Pin requests and its transitive dependencies to tested versions.

  • Record cryptographic hashes and install with hash enforcement, for example:

    bash
    python3 -m pip install --require-hashes -r requirements.txt
    
  • Generate and review hashes only from trusted artifacts.

  • Explicitly configure the trusted Python package index where appropriate.

  • Use an isolated virtual environment and avoid installing packages with administrative privileges.

  • Add automated dependency vulnerability and integrity monitoring.

Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Taint TrackingDirect Taint Flow, Variable-Mediated Taint Flow, Credential Exfiltration Chain
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (4)

Tainted flow: 'HEADERS' from os.environ.get (line 19, credential/environment) → requests.post (network output)

Critical
Category
Data Flow
Confidence
90% confidence
Finding

Credentials or environment variables flow to a network sink. This is a high-confidence indicator of credential exfiltration.

Content

Scanner excerpt · scripts/text.py (reported line 45)May include surrounding context.

python
if think:
        payload["thinking"] = {"type": "enabled", "budget_tokens": 18000}

    resp = requests.post(url, headers=HEADERS, json=payload, timeout=120)
    resp.raise_for_status()
    data = resp.json()

Undeclared Tool Scope

Medium
Category
MCP Least Privilege
Confidence
87% confidence
Finding

The skill documentation indicates capabilities that can access environment variables, perform network requests, and write output to files, but it does not declare any explicit tool scope or permissions. This creates an authorization and transparency gap: an agent or user may invoke the skill without clear constraints, increasing the risk of unintended secret exposure via MINIMAX_API_KEY, unreviewed outbound requests, or writing sensitive model output to disk.

Content

No source excerpt is available for this finding.

External Transmission

Medium
Category
Data Exfiltration
Confidence
80% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · scripts/text.py (reported line 45)May include surrounding context.

python
if think:
        payload["thinking"] = {"type": "enabled", "budget_tokens": 18000}

    resp = requests.post(url, headers=HEADERS, json=payload, timeout=120)
    resp.raise_for_status()
    data = resp.json()

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
93% confidence
Finding

The natural-language description, headings, and examples are presented only in Chinese, which can impose a language preference on users without an explicit opt-in or alternative. The policy scope specifically includes language or locale constraints across all file types.

Content

No source excerpt is available for this finding.