ClawHub
Back to skill

Security audit

Meyhem Search

Security checks for vulnerabilities and agentic risk

Overview

This is a straightforward web-search skill that clearly sends search queries to its external API and does not show hidden local access, persistence, or credential handling.

Install only if you are comfortable sending search queries, agent identifiers, and selected result URLs to api.rhdxm.com. Do not include secrets, proprietary text, private documents, or sensitive personal data in queries or agent_id values.

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
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (5)

Lp3

Medium
Category
MCP Least Privilege
Confidence
96% confidence
Finding
The skill clearly performs network operations to a third-party service but does not declare any tool scope or permissions metadata to make that capability explicit. This creates a transparency and governance gap: agents or users may invoke a skill that transmits data externally without an enforceable declared network permission boundary.

External Transmission

Medium
Category
Data Exfiltration
Content
## Quick Start (REST)

Full API docs: https://api.rhdxm.com/docs

```bash
curl -s -X POST https://api.rhdxm.com/search \
Confidence
50% confidence
Finding
Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

External Transmission

Medium
Category
Data Exfiltration
Content
Full API docs: https://api.rhdxm.com/docs

```bash
curl -s -X POST https://api.rhdxm.com/search \
  -H 'Content-Type: application/json' \
  -d '{"query": "YOUR_QUERY", "agent_id": "my-agent", "max_results": 5, "freshness": "hour"}'
```
Confidence
92% confidence
Finding
The HTTPS endpoint in the curl example represents actual outbound transmission of user-controlled content to a third-party search API. In the context of an agent skill, this is more sensitive because prompts, task context, or identifiers may be embedded into queries and unintentionally exfiltrated outside the local environment.

External Transmission

Medium
Category
Data Exfiltration
Content
Full API docs: https://api.rhdxm.com/docs

```bash
curl -s -X POST https://api.rhdxm.com/search \
  -H 'Content-Type: application/json' \
  -d '{"query": "YOUR_QUERY", "agent_id": "my-agent", "max_results": 5, "freshness": "hour"}'
```
Confidence
92% confidence
Finding
The HTTPS endpoint in the curl example represents actual outbound transmission of user-controlled content to a third-party search API. In the context of an agent skill, this is more sensitive because prompts, task context, or identifiers may be embedded into queries and unintentionally exfiltrated outside the local environment.

External Transmission

Medium
Category
Data Exfiltration
Content
## MCP

You can also connect via MCP at `https://api.rhdxm.com/mcp/` for richer integration.

## Data Transparency
Confidence
84% confidence
Finding
The MCP endpoint describes an additional remote integration channel to the same third-party service, expanding the skill's external communication surface. MCP-based integration may expose richer agent context than a simple search request, increasing the risk of unintended disclosure if not tightly scoped.

Static analysis

No suspicious patterns detected.