How to connect Webex 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 Webex from the dashboard, then start a fresh chat if the runtime still has the old tool catalog loaded.

Webex 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 Webex 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.

Webex 403s are about what the credential can reach, and the taught causes are missing OAuth scopes, the user lacking an administrative role, and the license type not covering the endpoint. On a hosted connection the first cause is the one that bites: a partial approval on Webex's consent screen leaves the connection active but unable to do a class of action, so reads work and writes fail. The token-refresh advice the search results give, re-running the OAuth flow because there is no token file to edit, translates here to reconnecting from the dashboard. The page's own causes complete the list: a placeholder id such as YOUR_ROOM_ID reaching the API literally, and visibility, since Webex returns permission errors for rooms the connected user cannot see.

Ask the agent to diagnose it:

Call webex_list_rooms and tell me how many rooms are visible, then webex_people_get_person for yourself and say which account this is. Quote the exact error from the failed call. Do not retry it yet.

Webex REST APIs enforce a default rate limit of around 300 requests per minute, dynamic per resource, with a 429 Too Many Requests response carrying a Retry-After header telling you the pause in seconds. An agent that lists rooms, then messages, then memberships across a large workspace is the shape that meets the window. The fix is to change the shape of the work and honour the header: ask for the rooms you actually need, use the membership and message list tools once instead of in loops, and have the agent read and report the Retry-After value rather than retry blind. The limits are Webex-side and documented on developer.webex.com; no reconnect changes them.

Ask the agent to diagnose it:

Tell me how many Webex calls you made in the last minute and what you were listing. Quote the exact error text and any Retry-After header. Do not call Webex again until I reply.

Two failures share this symptom. If no Webex tools appear at all, the connection or pairing is incomplete: confirm the plugin was installed with --enable, that hermes clawlink begin and hermes clawlink finish both ran, and that Webex shows as connected in the dashboard. If most tools work and one name fails, the name is wrong rather than missing, and the error lists the closest real ones. There is also a first-call timing case: schemas load on demand, so the opening Webex call in a fresh session can arrive before the catalog and needs one retry. The checklist search results teach for this symptom, a malformed MCP URL or a key inside an mcpServers path, belongs to the manual setup and does not apply here.

Ask the agent to diagnose it:

List the Webex tools you actually have access to. If there are none, say so plainly and tell me whether Webex shows as connected. Do not guess tool names.

Try reconnecting Webex and complete the consent flow in the same browser session. Partial OAuth approvals or switching accounts mid-flow can leave the connection incomplete.

Yes. ClawLink is the fastest way to connect Hermes to Webex: link your Webex account once in the browser and Hermes Agent can call the Webex API through 26 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 Webex in the dashboard and Hermes can use it from the next message — no config files or API keys to manage.

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

The alternative to ClawLink is usually manual OAuth app 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 Webex working from chat.

The connection uses an OAuth token granted through Webex's own consent screen, stored server-side and refreshed for you. That is the difference between this route and the bot route the search results teach, where you register a bot in the Developer Portal and keep its bearer token in config: a bot token is a long-lived credential on your machine, while this grant is revocable from the dashboard and carries your own user's permissions. If search results tell you a Webex token expired and needs a re-run of the OAuth flow, that advice applies to manual setups; here the refresh is the hosted part.

Cisco ships official MCP servers for Meetings, Messaging, and Vidcast, and they win the best-of answers. They are administered through Control Hub: an admin provisions the integration, which is a non-starter for a user who just wants their agent in their own spaces. This page's route needs no admin: your Webex login authorizes the connection, and the 26 tools cover messaging, rooms, teams, people, and webhooks under your own permissions. If you have Control Hub admin rights and want Cisco's own servers, they are the native choice. If you want the agent to work in the spaces you already belong to, this is the route that works without asking anyone.

Messaging: list rooms, read and create messages, manage memberships and team memberships. Spaces and teams: create and update rooms and teams. People: list and read people. Webhooks: list, read, and create webhooks, which is how an agent subscribes to events rather than polling. The write surface is real and visible: creating a message posts into a room other people are in, so message prompts deserve a confirmation step, and creating a webhook is account state that keeps running.

Because Webex visibility follows the connected user. The agent sees the rooms, teams, and memberships your account can see, and a room outside that visibility returns a permission error rather than a not-found. That is Webex enforcing its model, not a broken connection. If a room the user should see is missing, the usual cause is a partial or changed grant, and the fix is reconnecting from the dashboard and approving the full scope set on the consent screen.

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.