ClawHub
Back to skill

Security audit

trek-agent-control

Security checks for vulnerabilities and agentic risk

Overview

This skill connects an agent to the user's Trek travel account using a disclosed remote MCP service and CLI, with real but purpose-aligned access to read and update travel data.

Install only if you intend this agent to access and modify your Trek trips. Use a separate Trek Agent Key for this agent, store it in a secret manager when possible, revoke it if exposed, and review previews before allowing destructive, financial, reservation, proposal-decision, or bulk trip changes.

SkillSpector

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 (3)

External Transmission

Medium
Category
Data Exfiltration
Content
1. Never paste the `trek_` key into files, prompts, logs, screenshots, commits, or shell history intended for sharing.
2. Prefer the agent's native Streamable HTTP MCP support. Configure:
   - URL: `https://api.superd.fun/mcp`
   - Header: `Authorization: Bearer <user-key>`
3. Use the Trek CLI to install/sync this Skill, verify the connection, and diagnose failures:
Confidence
89% confidence
Finding
The skill instructs the agent to transmit a high-privilege `trek_` bearer token to a third-party remote MCP endpoint and also to install and use a remote CLI package. In context, this is the intended integration path, but it still creates a real security risk: a compromised endpoint, package source, or misconfigured agent could expose credentials and grant broad access to the user's Trek data.

External Transmission

Medium
Category
Data Exfiltration
Content
"mcpServers": {
    "trek": {
      "type": "streamable-http",
      "url": "https://api.superd.fun/mcp",
      "headers": {
        "Authorization": "Bearer ${TREK_MCP_TOKEN}"
      }
Confidence
95% confidence
Finding
The skill instructs agents to send an authorization bearer token to a remote third-party endpoint over Streamable HTTP MCP. Because the skill handles travel data and supports read/write synchronization, this creates a real external transmission path for sensitive itinerary data and credentials, and compromise or misuse of the remote service could expose or alter user data.

External Transmission

Medium
Category
Data Exfiltration
Content
- `TREK_CONFIG`: custom config path.
- `TREK_MCP_TOKEN`: override the stored key.
- `TREK_MCP_URL`: defaults to `https://api.superd.fun/mcp`.
- `TREK_MCP_TIMEOUT_MS`: per-request timeout, default `20000`.
- `TREK_MCP_RETRIES`: retry count for 429/502/503/504 and network errors, default `7`.
Confidence
93% confidence
Finding
The configuration exposes a default remote MCP URL and environment-variable overrides that route authenticated requests to an external service. In this skill context, those requests can include sensitive travel records and write operations, so the external transmission channel is security-relevant and could be abused if the endpoint, configuration, or runtime environment is compromised.

Static analysis

No suspicious patterns detected.