Back to skill

Security audit

Clawmart Skills Empire

Security checks for vulnerabilities and agentic risk

Overview

This package is mostly transparent marketing and template material, but some templates promote lead scraping, CRM export, trading integrations, API keys, and external alerts without enough safety guidance.

Review before installing or selling from this package. Treat the lead-generation template as handling regulated or sensitive contact data, and add consent, lawful-use, retention, and export controls. Store API keys outside plaintext config files, restrict Telegram/Discord or CRM tokens, and pin/audit dependencies before following the pip install instructions.

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
templates/trading-signals.md:17
Finding
Unpinned Third-Party Dependency Installation## Vulnerability Details **File Location**: `templates/trading-signals.md`, line 17 **Vulnerability Type**: Unpinned third-party dependencies **Risk Level**: Medium ```bash pip install MetaTrader5 pandas numpy ``` ### Technical Analysis The installation command resolves the latest available versions of three third-party packages without version constraints, cryptographic hashes, a lockfile, or an explicitly trusted package index. Consequently, the code installed by users can differ from the code that was available when the skill was reviewed. Python packages can execute package-controlled build or installation logic and later execute code when imported. If a dependency, its release process, or one of its transitive dependencies is compromised, following this instruction could introduce attacker-controlled code into the user's environment. This finding identifies a supply-chain weakness. The reviewed files do not establish that any named package is currently malicious. ### Attack Path 1. An attacker compromises a dependency release, maintainer account, distribution process, or transitive dependency. 2. The attacker publishes a malicious package version to the package index used by `pip`. 3. A user follows the documented installation command. 4. Because no versions or hashes are specified, `pip` may resolve and install the attacker-controlled release. 5. Malicious code executes during package installation, package import, or use of the trading functionality. ### Impact Assessment Malicious dependency code could run with the privileges of the user executing `pip` or importing the package. Depending on those privileges and the surrounding environment, this could permit access to local files, environment variables, API credentials, trading configuration, network resources, and any accounts accessible to that user. It could also modify files or establish additional malicious behavior. The scope is normally limited to the ...[truncated 221 chars]
Remediation
## Remediation Suggestions 1. Move dependencies into a reviewed lockfile or requirements file and pin exact versions. 2. Require cryptographic hashes, for example through `pip install --require-hashes -r requirements.txt`. 3. Pin and audit transitive dependencies, not only the three direct dependencies. 4. Configure an explicitly trusted package index or an internally controlled package mirror. 5. Install packages in a dedicated virtual environment or restricted container rather than a privileged system environment. 6. Run dependency vulnerability and provenance checks in CI before publishing updates. 7. Review dependency updates before changing pins, then regenerate and verify hashes. 8. Avoid running `pip` as root or with administrator privileges.
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 (4)

Ae1

High
Category
analysis-evasion
Content
3. Edit `SKILL.md` with your documentation
Confidence
100% confidence
Finding
Referenced artifact was not completely inspected

Missing User Warnings

Medium
Confidence
88% confidence
Finding
The skill advertises scraping, email finding/verification, and CRM export without any warnings or usage constraints around handling personal or business contact data. This can encourage collection, verification, and transfer of potentially sensitive data in ways that may violate privacy, consent, platform terms, or data-protection requirements.

Missing User Warnings

Medium
Confidence
91% confidence
Finding
The skill instructs users to configure API keys and advertises Telegram/Discord alerts, but it provides no warning about secure credential storage, least-privilege token use, or the fact that market data/signals may be sent to external services. This creates a real security/privacy risk because users may place secrets in plaintext config files and enable outbound notifications without understanding data exposure or account-abuse implications.

Natural-Language Policy Violations

Low
Confidence
86% confidence
Finding
Line L19 labels one skill as "Customer Support Bot - AI客服," mixing in a Chinese-language term without any indication that language or locale is optional or region-specific. This can conflict with a language/locale policy requiring user choice or documented justification for non-default language constraints.

Static analysis

No suspicious patterns detected.