ClawHub
Back to skill

Security audit

AnyoneCLI

Security checks for vulnerabilities and agentic risk

Overview

The skill is coherent, but it can publish permanent public profile data to a third-party service without a clear user-review checkpoint or deletion path.

Install only if you are comfortable with an agent helping create a public AnyoneCLI profile. Before any registration, review the exact JSON payload, remove secrets/private emails/internal URLs, and remember the artifact says profiles are public, permanent, and stored in a public GitHub repository with no edit/delete flow supported by this skill.

A.I.G

Vulnerability Patterns
  • 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
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs

SkillSpector

Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (6)

Autonomous Decision Making

Medium
Category
Excessive Agency
Content
OpenClaw / Claude Code skill for [AnyoneCLI](https://anyonecli.com).

One POST to register a terminal-style profile page. Instant. Free. No approval.

## Install
Confidence
87% confidence
Finding
The phrase 'No approval' promotes immediate registration without a review or confirmation checkpoint, which can enable autonomous or accidental external account/profile creation. In a skill used by AI agents, reducing human approval friction materially increases the chance of unintended publication of identity or profile information to a public external service.

Missing User Warnings

Medium
Confidence
93% confidence
Finding
The README encourages profile registration via an external API but does not warn users that their data will be transmitted off-platform to anyonecli.com. In an agent skill context, missing disclosure increases the risk of users or autonomous agents sending personal, organizational, or identifying information to a third party without informed consent.

Autonomous Decision Making

Medium
Category
Excessive Agency
Content
---
name: anyonecli
version: 1.0.0
description: "CLI identity for humans and AI agents. Register a terminal-style profile page at anyonecli.com with one POST — instant, free, no approval needed. Use when: creating a developer/agent profile, looking up someone by username via curl, self-registering agent identity. Triggers: 'register on anyonecli', 'create my CLI profile', '注册anyonecli', '创建开发者名片'. NOT for: editing existing profiles, on-chain identity, agent-to-agent messaging."
tags: ["developer-profile", "cli", "api", "identity", "agent-identity", "curl", "developer-tools"]
homepage: https://anyonecli.com
user-invocable: true
Confidence
87% confidence
Finding
The phrase 'no approval needed' encourages autonomous completion of a public registration action without emphasizing consent, review, or data sensitivity. In the context of an AI skill that can be invoked to create public identities, this increases the chance an agent performs irreversible external actions on behalf of a user without sufficiently informed authorization.

Missing User Warnings

Medium
Confidence
94% confidence
Finding
The registration example immediately instructs users or agents to POST profile data without placing a prominent warning that all submitted information becomes public, permanent, and stored in a public GitHub repository. In an agent setting, this raises the risk of oversharing personal, organizational, or sensitive identity data because the action is framed as simple and approval-free.

External Transmission

Medium
Category
Data Exfiltration
Content
## Register a Profile

```bash
curl -X POST https://anyonecli.com/api/register \
  -H "Content-Type: application/json" \
  -d '{
    "name": "my-agent",
Confidence
92% confidence
Finding
The skill contains a ready-to-run command that transmits user-supplied profile data to an external service, creating a direct exfiltration path for any information an agent includes in the payload. Because the skill is user-invocable and promotes self-registration for humans and AI agents, an assistant could autonomously send identifying or sensitive metadata off-platform without adequate review.

Natural-Language Policy Violations

Low
Confidence
86% confidence
Finding
The final line switches to Chinese without indicating that the skill supports multiple languages or that Chinese is required for a specific audience or compliance reason. This can violate a language/locale policy when a specific language is presented without user opt-in or documented justification.

Static analysis

No suspicious patterns detected.