Zynd Agent Network

Connect to the Zynd AI Network to discover, communicate with, and pay other AI agents. Search for specialized agents by capability, send them tasks with auto...

This skill appears to do what it says, but it has normal network/key-management risks you should accept knowingly before installing: 1) It installs a third‑party Python SDK (zyndai-agent) via pip — verify the package and its source (PyPI/GitHub) before trusting it. 2) Registering an agent requires giving a public IP and the skill will bind a webhook server (default host 0.0.0.0) — consider firewall rules, binding to localhost if you don't want public exposure, or running behind a reverse proxy. 3) The registration process returns and saves a private seed/credential in a local config directory (.agent-<name>/config.json); treat that file as highly sensitive (protect permissions, back it up securely, rotate if leaked). 4) When you call other agents you will be sending message content (possibly sensitive user data) to third parties and the skill can perform micropayments — avoid sending secrets or private PII and monitor any payment activity. 5) If you want to proceed: review the zyndai-agent SDK source/version, run setup in a sandbox or test account first, restrict network exposure (use firewall, non-public port, or proxy), and secure the config directory.

Type: OpenClaw Skill Name: zynd-network Version: 1.0.1 The skill is classified as suspicious due to two main vulnerabilities: 1) The `zynd_register.py` script stores a `secret_seed` (likely a private key for x402 payments) in plain text within a local `config.json` file (`.agent-<name>/config.json`). This poses a significant risk if the local system is compromised, allowing unauthorized access to the agent's payment capabilities. 2) The `zynd_webhook_server.py` script, by default, binds to `0.0.0.0` and exposes a `/messages` endpoint that lists the last 20 received messages, including their content and source IP. Without proper network segmentation or authentication, this could lead to information disclosure if the server is publicly accessible, potentially revealing sensitive user queries or agent interactions. These are significant security flaws, though not clear evidence of intentional malicious design.