Back to skill

Security audit

lgd-crypto-guard

Security checks for vulnerabilities and agentic risk

Overview

The skill itself is a local compliance checklist tool, but its install instructions ask users to run unpinned global or mutable remote installers.

Review the packaged script as the trusted artifact and avoid the documented unpinned install paths unless you can verify the package version, repository commit, and checksum. Treat the tool as an advisory compliance checklist, not legal or regulatory advice.

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:77
Finding
Unpinned Third-Party Installer and Mutable Remote Skill Source## Vulnerability Details **File Location**: `SKILL.md:77-81` **Vulnerability Type**: Supply-chain risk caused by unpinned third-party installation sources **Risk Level**: Medium ```bash npx skills add zhaoxinghua09-cell/agent-skills -g git clone https://github.com/zhaoxinghua09-cell/agent-skills.git cp -r agent-skills/skills/lgd-crypto-guard ~/.workbuddy/skills/ ``` ### Technical Analysis The documented installation process does not pin the npm CLI to an exact audited version. Consequently, `npx` may resolve, download, and execute a package version that differs from the version reviewed during this audit. The alternative installation procedure clones a mutable remote repository without specifying an audited commit, release tag, checksum, or cryptographic signature. Content obtained by future users can therefore differ from the audited project. Copying that content into an Agent skill directory may cause subsequently modified instructions or scripts to be trusted and invoked by the Agent. The `-g` option also requests global installation, increasing the potential scope of a compromised installer. No malicious behavior was found in the packaged `scripts/crypto_guard.py`; this finding concerns the unsafe installation and supply-chain process rather than the audited local implementation. ### Attack Path 1. An attacker compromises the referenced npm package, package publisher, GitHub account, or upstream repository. 2. The attacker publishes a malicious package version or modifies the repository's mutable default branch. 3. A user follows the documented `npx` or `git clone` installation instructions. 4. `npx` downloads and executes the current unpinned package, or Git retrieves modified Skill content that was not included in this audit. 5. In the npm path, attacker-controlled code can execute with the installing user's privileges. In the Git path, malicious scripts or instructions can become trusted after being copied into ...[truncated 866 chars]
Remediation
## Remediation Suggestions 1. Pin the npm CLI to an exact, reviewed version rather than relying on implicit current-version resolution: ```bash npx --yes skills@<audited-version> add zhaoxinghua09-cell/agent-skills ``` 2. Document and verify the expected npm package integrity hash and publisher identity. 3. Avoid global installation unless it is strictly required. Prefer a dedicated, least-privileged environment. 4. Pin repository installation to a reviewed commit: ```bash git clone https://github.com/zhaoxinghua09-cell/agent-skills.git cd agent-skills git checkout --detach <audited-commit-sha> ``` 5. Publish SHA-256 checksums or signed release artifacts and require users to verify them before copying or executing files. 6. Install only the specific audited Skill artifact rather than cloning and trusting the entire repository. 7. Review all downloaded scripts and Skill instructions before placing them in an Agent-controlled directory. 8. Add automated release controls that ensure documentation pins the installer version and source commit corresponding to each audited release.
Vulnerability Patterns
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (7)

Natural-Language Policy Violations

Medium
Confidence
88% confidence
Finding
The attestation is primarily written in Chinese across the file, with only a partial English title, and provides no indication that language selection is optional or user-driven. Under the policy rule for natural-language violations, this can constitute a language/locale constraint without opt-in or documented justification.

Natural-Language Policy Violations

Medium
Confidence
83% confidence
Finding
The natural-language instructions and operational guidance are presented only in Chinese, which can force a specific language on users without opt-in. The file includes an English display name, but it does not provide equivalent English instructions or state that the skill is intended only for a Chinese-language audience.

Vague Triggers

Medium
Confidence
89% confidence
Finding
The usage section says the skill should be proactively installed whenever the user is working on privacy computing, encryption, minimization AI applications, compliance checks, bidding materials, or regulatory response. These are broad scenario descriptions rather than specific invocation triggers or constrained activation rules, and no negative examples or exclusions are provided.

Rp1

Medium
Category
MCP Rug Pull
Confidence
84% confidence
Finding
The skill instructs users to run `npx skills` without pinning an exact package version or integrity source. This can cause execution of a newer or compromised package at install time, creating a supply-chain risk where arbitrary code may run on the user's machine.

Natural-Language Policy Violations

Medium
Confidence
95% confidence
Finding
This file contains natural-language instructions and descriptions that force a specific language/locale for users. The policy allows locale constraints only when user choice is offered or the restriction is clearly documented and justified, which is not present here.

Natural-Language Policy Violations

Medium
Confidence
93% confidence
Finding
The argparse description, help strings, and usage error message are user-facing natural language. Because they are only in Chinese and provide no language selection or justification, they violate the language/locale policy for natural-language content.

Natural-Language Policy Violations

Low
Confidence
90% confidence
Finding
The manifest uses Chinese-language category text ("AI 治理") while providing no indication that the skill is region-specific or that language/locale is user-selectable. Under the policy rule for language or locale constraints, this can be a natural-language policy issue when the locale is effectively fixed without opt-in or justification.

Static analysis

No suspicious patterns detected.