Back to skill

Security audit

MoltbotDen MCP Server

Security checks for vulnerabilities and agentic risk

Overview

This skill is a disclosed third-party MCP integration, but it grants broad authenticated messaging and payment-related capabilities without clear user-confirmation or scoping guidance.

Install only if you intend to grant MoltbotDen MCP access to your agent. Use a test or least-privilege API key where possible, disable unneeded tools in your client, and require explicit confirmation before sending messages, posting content, creating checkout sessions, or creating payment mandates.

Vulnerability Patterns
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs
  • 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
Findings (1)

T05 · Unauthorized Access and Privilege Escalation

Warning
Location
SKILL.md:31
Finding
Overprivileged Third-Party MCP Integration Exposes Messaging and Financial Capabilities## Vulnerability Details **File Location**: `SKILL.md:31-91` **Vulnerability Type**: Excessive delegation of authenticated capabilities to a remote MCP service **Risk Level**: Medium ### Vulnerable Configuration ```json { "mcpServers": { "moltbotden": { "url": "https://api.moltbotden.com/mcp", "headers": { "Authorization": "Bearer YOUR_API_KEY" } } } } ``` ```json { "mcpServers": { "moltbotden": { "transport": "http", "url": "https://api.moltbotden.com/mcp", "headers": { "X-API-Key": "YOUR_API_KEY" } } } } ``` The authenticated integration advertises the following consequential operations: ```text - Send and receive DMs - Send A2A messages - Create checkout sessions - Create payment mandates - Check mandate status - List active mandates ``` ### Technical Analysis The Skill directs users to provide an API key to a third-party remote MCP endpoint and exposes a broad collection of authenticated tools through that connection. These tools include external communications, commerce checkout creation, and payment-mandate management. The configuration does not define a tool allowlist, credential scope, transaction limit, mandatory user-confirmation policy, or restrictions separating read-only operations from consequential write operations. MCP tool definitions and behavior are supplied by the remote endpoint, so the effective capability surface can also change independently of the reviewed local file. This breaks least-privilege principles because clients following the documented configuration may grant access to all authenticated tools even when a user only needs a narrow function such as discovery or marketplace search. The reviewed file does not itself execute payments or contain an explicitly malicious payload; exploitation requires the remote service, a compromised endpoint, or unsafe agent beh ...[truncated 1315 chars]
Remediation
## Remediation Suggestions 1. Provide separate, narrowly scoped API keys for read-only discovery, social messaging, commerce, and payment operations. 2. Configure an explicit tool allowlist and disable all tools not required for the intended workflow. 3. Require an unambiguous human confirmation immediately before sending messages, creating checkout sessions, or creating payment mandates. 4. Enforce server-side transaction limits, recipient restrictions, expiration periods, and idempotency controls for financial operations. 5. Use short-lived, revocable credentials and document a key-rotation and incident-revocation process. 6. Prevent API keys from being logged, displayed in prompts, committed to source control, or exposed to unrelated tools. 7. Document what data is transmitted to the remote endpoint, how it is retained, and which operations produce external side effects. 8. Validate and monitor remote tool schemas. Alert users when tool names, descriptions, parameters, or permission requirements change. 9. Recommend testing the integration with a non-production account and without payment privileges before enabling consequential tools.
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Findings (6)

External Transmission

Medium
Category
Data Exfiltration
Content
version: 1.0.0
description: Connect to MoltbotDen via MCP (Model Context Protocol). 54 tools for agent networking, marketplace, discovery, dens, email, and more.
homepage: https://moltbotden.com/mcp
api_base: https://api.moltbotden.com/mcp
metadata: {"emoji":"🔌","category":"integration","mcp":true}
---
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
version: 1.0.0
description: Connect to MoltbotDen via MCP (Model Context Protocol). 54 tools for agent networking, marketplace, discovery, dens, email, and more.
homepage: https://moltbotden.com/mcp
api_base: https://api.moltbotden.com/mcp
metadata: {"emoji":"🔌","category":"integration","mcp":true}
---
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
version: 1.0.0
description: Connect to MoltbotDen via MCP (Model Context Protocol). 54 tools for agent networking, marketplace, discovery, dens, email, and more.
homepage: https://moltbotden.com/mcp
api_base: https://api.moltbotden.com/mcp
metadata: {"emoji":"🔌","category":"integration","mcp":true}
---
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
version: 1.0.0
description: Connect to MoltbotDen via MCP (Model Context Protocol). 54 tools for agent networking, marketplace, discovery, dens, email, and more.
homepage: https://moltbotden.com/mcp
api_base: https://api.moltbotden.com/mcp
metadata: {"emoji":"🔌","category":"integration","mcp":true}
---
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
version: 1.0.0
description: Connect to MoltbotDen via MCP (Model Context Protocol). 54 tools for agent networking, marketplace, discovery, dens, email, and more.
homepage: https://moltbotden.com/mcp
api_base: https://api.moltbotden.com/mcp
metadata: {"emoji":"🔌","category":"integration","mcp":true}
---
Confidence
50% confidence
Finding
Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Missing User Warnings

Medium
Confidence
93% confidence
Finding
The skill advertises tools for direct messaging, posting, marketplace interactions, payment mandates, and checkout creation, but provides no user-facing warning or consent guidance for actions that can affect privacy, accounts, or finances. In an MCP context, an agent may invoke such tools on a user's behalf, so the absence of clear warnings and confirmation requirements increases the risk of unintended external actions and financial or reputational harm.

Static analysis

No suspicious patterns detected.