Back to skill

Security audit

AI IT Helpdesk

Security checks for vulnerabilities and agentic risk

Overview

The skill describes a plausible IT helpdesk tool, but installation runs mutable, unreviewed code from GitHub, so users should review it before installing.

Install only after reviewing the referenced GitHub repository and its dependencies, preferably pinned to a specific commit and run in an isolated virtual environment or container. Confirm how the helpdesk app authenticates users, protects ticket and knowledge-base data, and controls ticket-changing actions before using it with real business data.

Vulnerability Patterns
  • Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
  • 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
  • Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs
Findings (1)

T03 · Remote Payload Retrieval and Execution

Error
Location
SKILL.md:41
Finding
Unpinned Remote Payload Retrieval and Execution## Vulnerability Details **File Location**: `SKILL.md`, lines 41-44 **Vulnerability Type**: Remote payload retrieval and execution **Risk Level**: High ```bash git clone https://github.com/openclaw-skills/ai-intelligent-helpdesk cd ai-intelligent-helpdesk pip install -r requirements.txt python app.py ``` ### Technical Analysis The installation procedure retrieves the current default branch of a remote Git repository without pinning it to a reviewed commit hash or verifying its integrity. It subsequently installs packages from a remotely supplied `requirements.txt` file and executes the remotely supplied `app.py`. Neither `requirements.txt` nor `app.py` is present in the audited package, so their behavior cannot be inspected or verified as part of this audit. The effective executable payload can change after the Skill has been reviewed because the remote repository is mutable. Although GitHub is a legitimate hosting service, it does not establish the trustworthiness or immutability of repository content. This creates a supply-chain execution channel: compromise of the repository, its maintainer account, or its dependencies could cause future users to install or execute attacker-controlled code. ### Attack Path 1. An attacker compromises the referenced repository or a maintainer account, or otherwise gains the ability to alter its default branch. 2. The attacker modifies `app.py`, `requirements.txt`, or a dependency installation hook to contain malicious code. 3. A user follows the documented installation instructions. 4. `git clone` downloads the modified content without checking a trusted commit or signature. 5. `pip install -r requirements.txt` may execute malicious package build or installation code. 6. `python app.py` directly executes the downloaded application with the privileges of the installing user. ### Impact Assessment Successful exploitation can provide arbitrary code execution under the installin ...[truncated 585 chars]
Remediation
## Remediation Suggestions 1. Include the complete executable source and dependency manifest in the Skill package so that the reviewed artifact is the artifact users execute. 2. If remote retrieval is necessary, check out an explicitly reviewed commit hash rather than the mutable default branch. 3. Publish signed releases and verify the release signature or a trusted cryptographic checksum before installation or execution. 4. Pin every Python dependency to an exact version and use a hash-locked requirements file, such as one installed with `pip --require-hashes`. 5. Review direct and transitive dependencies for malicious packages, typosquatting, known vulnerabilities, and unsafe build hooks. 6. Install dependencies inside an isolated virtual environment or container using a non-privileged account. 7. Avoid running installation or application commands as root, and restrict filesystem, credential, and network access to the minimum required. 8. Add automated integrity and provenance checks to ensure that the downloaded source exactly matches the version that passed security review.
Vulnerability Patterns
  • 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
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (2)

Natural-Language Policy Violations

Medium
Confidence
93% confidence
Finding
The manifest description and the main documentation are entirely in Chinese, with no indication that language selection is optional or that the skill is intentionally limited to Chinese-speaking users. This can violate a language/locale policy when a skill implicitly forces a specific language without user opt-in or justification.

Vague Triggers

Medium
Confidence
95% confidence
Finding
The description "AI intelligent ai-intelligent-helpdesk" is too generic to establish a clear activation boundary or user expectation for the skill. In an enterprise IT helpdesk context, vague scope increases the chance of unintended invocation or overbroad handling of sensitive support tasks, which can lead to misuse, privilege overreach, or unsafe automation decisions.

Static analysis

No suspicious patterns detected.