OpsBuddy

• Outcome: Pass
• Latest audit: Sep 16, 2026 · 10:43 AM UTC
• Version: 1.0.9

Overview

This is a documentation-only setup skill for a third-party OpsBuddy MCP connection, with sensitive credential and gateway risks disclosed to the user.

Install only if you trust the OpsBuddy portal and gateway with operational telemetry and any monitoring credentials you configure there. Use read-only, least-privilege provider credentials, inspect the downloaded MCP config before enabling it, keep the bearer token out of chat, and revoke or rotate it if exposed.

A.I.G

Vulnerability Patterns

• Insecure Dependencies: Introduces malicious components through unsafe dependency sources
• Skill Instruction Hijacking: Alters the agent's session goals or safety constraints when the skill loads
• Agent Memory Poisoning: Writes attacker-controlled rules into memory that affect later sessions
• Remote Payload Retrieval and Execution: Fetches external code whose behavior can change after review
• Embedded Malicious Code: Ships malicious scripts inside the skill and executes them locally

Findings (1)

T08 · Insecure Dependencies

Warning

• Location: SKILL.md:44

Finding

Unverified Third-Party MCP Gateway Handles Operational Credentials and Telemetry

Content

View full analysis

**⚠️ Sensitive-credential notice**: The downloaded `mcp-config.json` contains your **API Key (plaintext credential)**. Keep it out of chat. You enter it only inside the OpenClaw MCP config (e.g. `~/.openclaw/openclaw.json`), not by pasting it to me. If the Key leaks, revoke it on the portal and re-apply immediately. > **🔒 Data-flow disclosure**: The OpsBuddy MCP connection points to the `https://ywdz.lxiai.com/` gateway. Your monitoring-platform credentials (Signoz / Tencent Cloud / Aliyun / Nightingale) are configured by you on that portal and transit through this gateway. This skill never receives or stores those credentials. > **Core Principle**: This skill only *explains* how to connect. It never receives an API Key, never asks the user to paste a config JSON, and never writes MCP config files. Platform connections, asset management and all other ops operations are done by the user on t ...[truncated 2840 chars]

Remediation

View remediation

SkillSpector

Vulnerability Patterns

• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access
• Excessive Agency: Unrestricted Tool Access, Autonomous Decision Making, Scope Creep
• Rogue Agent: Self-Modification, Session Persistence
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands

Findings (8)

MCP Config Access

High

• Category: Agent Snooping
• Confidence: 90% confidence

Finding

Skill accesses MCP server configuration files (mcp.json). MCP configs contain server URLs, authentication tokens, and tool definitions — reading them allows the skill to discover and potentially abuse other tool integrations.

Content

Scanner excerpt · SKILL.md (reported line 8)May include surrounding context.

md

slogan: An AI Agent Swarm That Autonomously Discovers Issues and Proactively Repairs Faults
license: Commercial / Commercial License (Non-Open-Source)
pricing: Key is currently free to apply (self-registration → auto-approval)
token: An API Key (MCP token) is required. Self-register → auto-approved → log in to the portal, go to "My Applications" and download the MCP config JSON (contains the API Key). No default key.
requires_api_key: true   # This Skill requires an API Key (the OpsBuddy MCP token) to work
---

MCP Config Access

High

• Category: Agent Snooping
• Confidence: 90% confidence

Finding

Skill accesses MCP server configuration files (mcp.json). MCP configs contain server URLs, authentication tokens, and tool definitions — reading them allows the skill to discover and potentially abuse other tool integrations.

Content

Scanner excerpt · SKILL.md (reported line 44)May include surrounding context.

md

slogan: An AI Agent Swarm That Autonomously Discovers Issues and Proactively Repairs Faults
license: Commercial / Commercial License (Non-Open-Source)
pricing: Key is currently free to apply (self-registration → auto-approval)
token: An API Key (MCP token) is required. Self-register → auto-approved → log in to the portal, go to "My Applications" and download the MCP config JSON (contains the API Key). No default key.
requires_api_key: true   # This Skill requires an API Key (the OpsBuddy MCP token) to work
---

MCP Config Access

High

• Category: Agent Snooping
• Confidence: 90% confidence

Finding

Skill accesses MCP server configuration files (mcp.json). MCP configs contain server URLs, authentication tokens, and tool definitions — reading them allows the skill to discover and potentially abuse other tool integrations.

Content

Scanner excerpt · SKILL.md (reported line 63)May include surrounding context.

md

slogan: An AI Agent Swarm That Autonomously Discovers Issues and Proactively Repairs Faults
license: Commercial / Commercial License (Non-Open-Source)
pricing: Key is currently free to apply (self-registration → auto-approval)
token: An API Key (MCP token) is required. Self-register → auto-approved → log in to the portal, go to "My Applications" and download the MCP config JSON (contains the API Key). No default key.
requires_api_key: true   # This Skill requires an API Key (the OpsBuddy MCP token) to work
---

MCP Config Access

High

• Category: Agent Snooping
• Confidence: 90% confidence

Finding

Skill accesses MCP server configuration files (mcp.json). MCP configs contain server URLs, authentication tokens, and tool definitions — reading them allows the skill to discover and potentially abuse other tool integrations.

Content

Scanner excerpt · SKILL.md (reported line 66)May include surrounding context.

md

slogan: An AI Agent Swarm That Autonomously Discovers Issues and Proactively Repairs Faults
license: Commercial / Commercial License (Non-Open-Source)
pricing: Key is currently free to apply (self-registration → auto-approval)
token: An API Key (MCP token) is required. Self-register → auto-approved → log in to the portal, go to "My Applications" and download the MCP config JSON (contains the API Key). No default key.
requires_api_key: true   # This Skill requires an API Key (the OpsBuddy MCP token) to work
---

Credential Access

High

• Category: Privilege Escalation
• Confidence: 70% confidence

Finding

Code accesses credential files (SSH keys, AWS credentials, etc.). This could indicate credential theft attempts.

Content

Scanner excerpt · SKILL.md (reported line 14)May include surrounding context.

md

# OpsBuddy Connection Skill

Helps the user register the "OpsBuddy" MCP connection into OpenClaw. **This Skill requires an API Key** (OpsBuddy access token / MCP Token); there is no default key and the user must obtain it on their own.

## Capability Overview (when the user asks "What are you?" / "What can you do?", output this section)

Autonomous Decision Making

Medium

• Category: Excessive Agency
• Confidence: 85% confidence

Finding

Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.

Content

Scanner excerpt · SKILL.md (reported line 8)May include surrounding context.

md

slogan: An AI Agent Swarm That Autonomously Discovers Issues and Proactively Repairs Faults
license: Commercial / Commercial License (Non-Open-Source)
pricing: Key is currently free to apply (self-registration → auto-approval)
token: An API Key (MCP token) is required. Self-register → auto-approved → log in to the portal, go to "My Applications" and download the MCP config JSON (contains the API Key). No default key.
requires_api_key: true   # This Skill requires an API Key (the OpsBuddy MCP token) to work
---

Autonomous Decision Making

Medium

• Category: Excessive Agency
• Confidence: 80% confidence

Finding

Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.

Content

Scanner excerpt · SKILL.md (reported line 40)May include surrounding context.

md

- **Trend comparison**: ask "How is health vs. yesterday" / "What's the 7-day / 30-day trend" → compares alerts and health across time windows.
- **Alert root-cause summary**: ask "My user-center page is laggy — analyze all possible software & hardware causes" → automatically traces the dependent database, middleware, hosts and business systems to pinpoint software/hardware root causes, giving a root-cause digest with recommendations instead of listing raw alerts.
> These showcase the value better than "check a metric" — guide users to ask them after connecting.
**⚠️ How to connect: this skill does NOT handle the connection for you — you MUST guide the user to complete it themselves on the portal / in OpenClaw. Never ask the user to paste a config JSON containing an API Key into the chat, and never write any MCP config file on the user's behalf.**

**Connection steps (user does it themselves):**

Session Persistence

Medium

• Category: Rogue Agent
• Confidence: 60% confidence

Finding

Skill establishes unauthorized persistence across sessions via cron jobs, startup scripts, or state files. Session persistence allows an attacker to maintain access beyond the current interaction.

Content

Scanner excerpt · SKILL.md (reported line 40)May include surrounding context.

md

- **Trend comparison**: ask "How is health vs. yesterday" / "What's the 7-day / 30-day trend" → compares alerts and health across time windows.
- **Alert root-cause summary**: ask "My user-center page is laggy — analyze all possible software & hardware causes" → automatically traces the dependent database, middleware, hosts and business systems to pinpoint software/hardware root causes, giving a root-cause digest with recommendations instead of listing raw alerts.
> These showcase the value better than "check a metric" — guide users to ask them after connecting.
**⚠️ How to connect: this skill does NOT handle the connection for you — you MUST guide the user to complete it themselves on the portal / in OpenClaw. Never ask the user to paste a config JSON containing an API Key into the chat, and never write any MCP config file on the user's behalf.**

**Connection steps (user does it themselves):**