Back to skill

Security audit

Ai Intelligent Attendance Management

Security checks for vulnerabilities and agentic risk

Overview

The skill describes a plausible attendance-management tool, but its install steps run unreviewed code from a mutable external GitHub repository.

Review this before installing. The attendance-management purpose is coherent, but do not run the documented git clone, pip install, or python app.py steps unless you have audited the referenced repository at a specific commit and its dependencies. Treat any deployment as handling sensitive employee data, and require access controls, logging, retention limits, and isolation from unnecessary credentials or private files.

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:36
Finding
Mutable Remote Payload Retrieval and Execution## Vulnerability Details **File Location**: `SKILL.md`, lines 36-40 **Vulnerability Type**: Remote payload retrieval and execution **Risk Level**: High The installation instructions retrieve an unpinned external repository, install its unaudited dependencies, and execute its application: ```bash git clone https://github.com/openclaw-skills/ai-intelligent-attendance-management cd ai-intelligent-attendance-management pip install -r requirements.txt python app.py ``` ### Technical Analysis The audited package contains only `SKILL.md` and `skill.json`; it does not include the referenced `app.py`, `requirements.txt`, or any application implementation. The effective executable payload therefore resides in an external Git repository and can change after this Skill has been reviewed. The `git clone` command does not pin a commit hash or verify a signed release. Running `pip install -r requirements.txt` may execute package installation hooks, while `python app.py` directly executes the retrieved application. Neither the dependency identities and versions nor the application's behavior can be verified from the supplied artifact. ### Attack Path 1. An attacker compromises the referenced repository, a maintainer account, or an included Python dependency. 2. The attacker modifies `app.py`, `requirements.txt`, or a dependency installation hook to contain an unauthorized payload. 3. A user follows the documented installation procedure without checking out a trusted commit or verifying artifact signatures. 4. `pip install` executes malicious package build or installation logic, or `python app.py` executes the altered application. 5. The payload operates with the privileges and data access of the user who ran the commands. ### Impact Assessment Successful exploitation can provide arbitrary code execution under the invoking user's account. The resulting scope may include access to files, environment variables, credentials, networ ...[truncated 603 chars]
Remediation
## Remediation Suggestions - Include the complete application source and dependency manifest in the reviewed Skill package. - Replace the mutable repository clone with a checkout pinned to a reviewed full commit hash or a cryptographically signed release. - Verify the repository identity, release signature, and downloaded artifact checksum before installation or execution. - Use a lock file containing exact dependency versions and cryptographic hashes, such as a hash-locked requirements file installed with `pip --require-hashes`. - Review direct and transitive dependencies for dependency confusion, typosquatting, known vulnerabilities, and unexpected installation hooks. - Run installation and application startup in an isolated, least-privileged environment without unnecessary credentials, sensitive filesystem access, or unrestricted network access. - Do not execute `app.py` until its contents and all imported local modules have been audited. - Ensure the packaged and externally published versions are reproducible from reviewed source.
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 (3)

Missing User Warnings

Medium
Confidence
93% confidence
Finding
The skill describes handling attendance, leave, and payroll-linked workflows, which inherently involve sensitive personal and employment data, but it provides no warnings or safeguards about privacy, access control, retention, or compliance obligations. This omission can lead users to deploy or use the skill without understanding the risks of exposing employee records, leave details, or salary-related information.

Natural-Language Policy Violations

Low
Confidence
77% confidence
Finding
The natural-language description and headings are entirely in Chinese, which can amount to forcing a specific language without user opt-in. The file does not indicate that the skill is region-specific or provide any option for users to choose another language.

Vague Triggers

Low
Confidence
95% confidence
Finding
The manifest description is overly generic and does not clearly constrain what the skill is supposed to do or when it should activate. In agent ecosystems, vague descriptions can broaden trigger scope, increase the chance of unintended invocation, and make review and policy enforcement harder.

Static analysis

No suspicious patterns detected.