How to connect Ngrok to Hermes Agent

Pairing is a two-step handshake: run hermes clawlink begin, approve the link in your browser, then run hermes clawlink finish. If you ran finish before approving, or the approval link expired, run hermes clawlink begin again to get a fresh link. Confirm the plugin was installed with --enable, then verify with hermes clawlink test.

Reconnect Ngrok from the dashboard, then start a fresh chat if the runtime still has the old tool catalog loaded.

Ngrok tool schemas load on demand the first time a tool runs and are cached after that, so this error usually clears on its own: wait a few seconds and retry the same request. If every Ngrok call keeps failing with it in a fresh chat, reconnect from the dashboard, and contact support if it still persists — that pattern points to a configuration problem on our side, not something you can fix by reconnecting again.

Two usual causes. The connected account may not have access to the specific workspace, inbox, store, or project in the request — check that first. If access looks right, the agent may have sent a placeholder value (like "YOUR_ID" or an example id from documentation) instead of a real one: ask it to run a list or search tool first, then retry the action with a real id from those results. Most failures at this stage are one of these two, not ClawLink bugs.

"Connected" means a key was accepted at setup, so start with the key. If it was deleted or rotated in ngrok afterwards, every call fails from that point and re-pasting the current key from the dashboard is the fix — a total failure across all tools is the signature. If some calls work and others do not, the key is alive and you are looking at what it is permitted to do: ngrok API keys carry the account's access, so a key from an account without rights to a resource returns a permission error for that resource while everything else behaves. The third cause is scope of account — ngrok resources belong to an account, so an endpoint or domain id from a different account fails the same way, as does a placeholder id copied from documentation.

Ask the agent to diagnose it:

Call ngrok_get_credentials and list what this connection can see. Then quote the exact error from the failed call and tell me which resource id it used. Do not retry yet.

Correct, and it is not confused — nothing here starts a tunnel from your machine, because that requires a local process and this integration is the management API. ngrok_create_endpoint configures an endpoint in your ngrok account; it does not open a connection from your laptop. If you need traffic reaching a locally running Hermes, run the ngrok CLI alongside it. If the agent invents a name like ngrok_start_tunnel, that name has never existed and the error will list the closest real tools.

Ask the agent to diagnose it:

List every ngrok tool you actually have access to and tell me plainly whether any of them can start a tunnel from this machine. Do not guess at names.

Two problems that present the same way. If Hermes shows no ngrok tools at all, the pairing or the connection is incomplete: check that the plugin was installed with --enable, that hermes clawlink begin and hermes clawlink finish both completed, and that ngrok shows as connected in the dashboard. If most tools work and one fails, that name is wrong rather than missing. There is also a first-call timing case: schemas load on demand, so the opening ngrok call in a fresh session can arrive before the catalog and just needs a retry. Editing a client config file, which is what search results for this symptom usually suggest, is not part of this setup.

Ask the agent to diagnose it:

List the ngrok tools you have access to. If there are none, say so. If there are, tell me which one lists endpoints and use that exact name.

Double-check that the API key for Ngrok has the right scopes or account access. A valid key can still be too limited for some reads or writes.

Yes. ClawLink is the fastest way to connect Hermes to Ngrok: link your Ngrok account once in the browser and Hermes Agent can call the Ngrok API through 102 ready-made tools — no custom code or token handling.

Install the plugin with hermes plugins install ClawLink-HQ/hermes-plugin --enable, then pair once: run hermes clawlink begin, approve the link in your browser, and run hermes clawlink finish. Connect Ngrok in the dashboard and Hermes can use it from the next message — no config files, and the Ngrok key you paste is stored server-side instead of in your environment.

About two minutes. Sign in, click Connect next to Ngrok in the dashboard, authenticate, and Hermes Agent can use it from the next chat message.

The alternative to ClawLink is usually manual API key setup plus your own token handling, permission troubleshooting, and tool plumbing for Hermes Agent. That is fine if you want to build and maintain the integration yourself. Most teams just want Ngrok working from chat.

No, and this is the distinction worth getting straight first. Starting a tunnel — what ngrok now calls an agent endpoint — from your own machine is the ngrok CLI's job, because something has to be listening locally for traffic to be forwarded to. This integration talks to ngrok's management API instead: it can create and configure endpoints, reserved domains, edges and their route modules, IP restriction policies, credentials, and API keys in your account, and read what already exists. Think of it as the ngrok dashboard as tools, rather than as a replacement for the CLI.

They solve unrelated problems that happen to share a vendor and a marketing phrase. ngrok's AI Gateway sits in front of model providers and routes LLM requests through a single key — it is about your agent's traffic to models. This integration is about your ngrok account itself: what endpoints exist, which domains are reserved, what the IP restrictions are, who holds credentials. If you searched for running an agent without managing API keys and landed on the gateway, that is a reasonable answer to a different question.

An ngrok API key, created in your ngrok dashboard, pasted once into the hosted setup page. It carries the access of the account it belongs to, so the agent can see and change what that account can. Because it is held server-side rather than in an environment variable, rotating it in ngrok means updating one place, and removing the connection from the ClawLink dashboard cuts access immediately rather than whenever someone cleans up a machine.

Auditing and changing account configuration without opening the dashboard. Asking which endpoints are live and what they point at, checking which reserved domains exist before someone claims another, reviewing IP restriction policies, and finding stale API keys and credentials that should have been revoked. The last one is the most common real use: ngrok accounts accumulate keys, and an agent that can enumerate them turns a periodic chore into a question you ask in chat.

Pairing is a two-step handshake: run hermes clawlink begin, approve the link in your browser, then run hermes clawlink finish. If you ran finish before approving, or the approval link expired, run hermes clawlink begin again to get a fresh link. Confirm the plugin was installed with --enable, then verify with hermes clawlink test.