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

Freshdesk 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 Freshdesk 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.

Freshdesk answers with permission errors in two situations, and the diagnosis order matters. First, the connected user's role and scope: an agent key whose Freshdesk user lacks access to a ticket, group, or company returns 403 for exactly those objects, and Freshdesk's own support article names this error class as "you are not allowed to access this page". Second, the arguments: a placeholder or invented ticket or contact id returns the same permission-shaped error because Freshdesk hides what the account cannot see. Check the id the agent sent before reissuing credentials, and only reconnect after both are ruled out.

Ask the agent to diagnose it:

Call freshdesk_currently_authenticated_agent and freshdesk_get_account. Tell me which Freshdesk user this connection acts as and which plan the account is on, then quote the exact error from the failed call. Do not retry it yet.

Freshworks documents rate limits of 1000 requests per hour on higher plans and per-minute limits on lower ones, and a bulk update of many tickets is the traffic shape that meets them: the API accepts the call and the allowance drains in one sweep. A 429 is the hour being spent, so retrying immediately is wasted traffic. Change the shape of the work: split a large update into batches across turns, and have the agent report how many requests it has made when it hits a 429 rather than retrying in a loop.

Ask the agent to diagnose it:

Tell me how many Freshdesk requests you have made in the last hour and what the last bulk call touched. Quote any rate-limit text from the last response. Do not call Freshdesk again until I reply.

Two different failures. If Hermes shows no Freshdesk tools at all, the connection or the pairing is incomplete: confirm the plugin was installed with --enable, that both hermes clawlink begin and hermes clawlink finish ran, and that Freshdesk shows as connected in the dashboard, then start a fresh chat. If most tools work and a single name fails, the name is wrong rather than missing, and the error lists the closest real ones. Search results teach "verify your API key and domain" for this symptom, which is the self-hosted version of the same failure; here the key is already verified at setup, so the name check is the relevant one. There is also a first-call timing case unique to this setup: schemas load on demand, so the opening Freshdesk call in a fresh session can arrive before the catalog and needs one retry.

Ask the agent to diagnose it:

List the Freshdesk tools you actually have access to. If there are none, say so plainly. If there are, tell me which one searches tickets and use that exact name.

Double-check that the API key for Freshdesk 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 Freshdesk: link your Freshdesk account once in the browser and Hermes Agent can call the Freshdesk API through 178 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 Freshdesk in the dashboard and Hermes can use it from the next message — no config files, and the Freshdesk key you paste is stored server-side instead of in your environment.

About two minutes. Sign in, click Connect next to Freshdesk 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 Freshdesk working from chat.

Because Freshdesk's API is key-authenticated, and an honest page says so. The key is entered once on the hosted setup page and held server-side; the agent never sees it, it is not written to any .env or config file on your machine, and it can be revoked from the dashboard at any time. The AI Overview that quotes ClawLink calls this "one-click hosted OAuth", which is the same flow described loosely: the key is the credential, the hosting is what removes the setup. The claim that matters, and the one the winning citations repeat, is that no configuration files are involved and nothing runs locally.

The ticket surface is the core: add notes and watchers, bulk-update tickets, create canned responses, and search tickets, contacts, companies, and agents. The read side covers agents and their availability, business hours, email mailboxes, discussion topics, and the account's company and contact fields. The writes to be deliberate about are notes on tickets, which customers can see depending on your Freshdesk settings, and bulk updates, which touch many tickets in one call.

Separate. Freshworks sells its own in-product AI Agent Studio (an Omni-plan feature) and an API Actions library, and those are what searches for "Freshdesk AI agent without API key" surface first. They are native to Freshdesk and configure through the product. This integration is the external-agent route: Hermes is the agent, Freshdesk is one connected account, and the same 178 tools are callable alongside your other integrations.

No, and the comparison is worth being exact about. Freshdesk's official MCP server is in early access, targets Enterprise plans, and still authenticates with an API key, so it does not remove the setup this page removes: the key still has to be created and stored somewhere you run a server. This integration holds the key for you, exposes 178 tools from chat, and needs nothing running. If you are on a plan where the official EAP is available and want a vendor-managed MCP endpoint, that is a real option; for an agent that works across apps, the hosted route is the one without the server.

Judge it by the mechanics. The credential is an API key held server-side, never shown to the agent and never written to your machine, and revoking it in the dashboard kills access immediately. The permission question is a Freshdesk account question: connect a key whose user has the roles the agent should act with, and check the "you are not allowed to access this page" error class when a tool fails, since Freshdesk answers restricted objects with permission errors rather than not-found. For writes like notes and bulk updates, tell the agent to draft before it changes anything.

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.