mcpc-mcp-client

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (high risk: 0.75). The skill’s runtime workflow can ingest outsider-authored free text from MCP server responses (e.g., tools-call, resources-read, prompts-get, tasks-result), which are not authored by the operating user and are fed into the agent/LLM context via the command output.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.90). The skill explicitly connects to external MCP servers at runtime (e.g., mcpc connect mcp.apify.com and mcpc connect https://mcp.example.com), which will fetch remote prompts/tools/resources from those hosts that can directly control agent instructions and behavior.

MEDIUM W013: Attempt to modify system services in skill instructions.

• Attempt to modify system services in skill instructions detected (high risk: 0.80). The skill includes explicit privileged commands (e.g., "sudo apt-get install libsecret-1-0 gnome-keyring" and global npm installs) that instruct the agent/user to obtain elevated privileges and modify the host system.