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

New Relic 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 New Relic 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.

Start with the key rather than the connection, because this is a key-based integration and "connected" only means a key was accepted at setup time. If the key was rotated or deleted in New Relic afterwards, every call fails from that moment and re-pasting the current key from the dashboard is the fix. If only some calls fail, the key is alive and you are looking at permissions: New Relic scopes access by the user the key belongs to, so a key from a read-only user queries fine and fails on anything that creates a policy or a key. The third cause is account selection — New Relic organisations can hold several accounts, and a key tied to one account returns permission errors for entities that live in another, which looks like a broken integration but is not.

Ask the agent to diagnose it:

Call new_relic_fetch_your_org_id and new_relic_get_applications, and tell me which org and account this connection can actually see. Then quote the exact error from the failed call. Do not retry it yet.

Usually the account or the window rather than the query. NRQL runs against a specific account, so a query issued against the wrong one in a multi-account organisation returns an empty result rather than an error. The second cause is the time window: NRQL defaults to a recent window unless a SINCE clause is given, so a question about last month returns nothing unless the agent says so explicitly. The third is retention — data older than your plan's retention period is genuinely gone and no query will find it. Ask the agent to show you the NRQL it is about to run, including the SINCE clause, before you debug anything else.

Ask the agent to diagnose it:

Show me the exact NRQL you ran, including the SINCE clause and which account it targeted. Then re-run it with SINCE 7 days ago and tell me what changes.

Two different problems. If Hermes has no New Relic tools at all, the connection or the pairing is incomplete — confirm the plugin was installed with --enable, that hermes clawlink begin and hermes clawlink finish both ran, and that New Relic shows as connected in the dashboard. If most tools work and one name fails, that name is wrong rather than missing and the error lists the closest real ones. There is also a timing case specific to this setup: schemas load on demand, so the first New Relic call in a fresh session can arrive before the catalog and simply needs a retry. None of these are fixed by editing a client config file, which is what most search results for this symptom assume.

Ask the agent to diagnose it:

List the New Relic tools you actually have access to. If there are none, say so plainly. If there are, tell me which one runs an NRQL query and use that exact name.

Double-check that the API key for New Relic 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 New Relic: link your New Relic account once in the browser and Hermes Agent can call the New Relic API through 158 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 New Relic in the dashboard and Hermes can use it from the next message — no config files, and the New Relic key you paste is stored server-side instead of in your environment.

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

Not natively, and that is the answer most sources stop at. Hermes ships without New Relic tools, so they have to come from somewhere — the usual suggestions are running an MCP server yourself and pointing Hermes at it, or wiring the New Relic API in by hand. Installing the ClawLink plugin gives Hermes the tools directly: one plugin install, one pairing handshake, one key pasted in the browser, and the tools appear in chat. Nothing runs on your machine afterwards.

New Relic issues several kinds of key for different jobs, and the one that matters here is a User key, which is what the REST and NerdGraph APIs accept for querying and managing your account. It goes into the hosted setup page once, and then lives server-side rather than in an environment variable or a config file. The practical consequences are worth knowing: if you rotate the key in New Relic you update it in one place, and if you remove the connection from the ClawLink dashboard the agent loses access immediately rather than whenever someone remembers to clean up a .env.

Yes, and it is the single most useful tool in the set. new_relic_execute_nrql_query runs NRQL against your account, which is how you get from a vague question to an actual number — error rates over a window, slowest transactions, throughput by host. The rest of the toolset covers the things around it: applications and their metrics, alert policies and conditions, violations, dashboards and widgets, notification channels and workflows, tags, and entity search. Ask for NRQL when you want data, and the entity or alert tools when you want to change configuration.

Yes, and that is the part to be deliberate about. The toolset can create alert policies and NRQL conditions, add notification channels to policies, and build AI workflows and destinations — real changes to whether your team gets paged. It can also create API access keys, which is a genuinely privileged action. The agent inherits whatever your New Relic user can do, so if it should not be able to touch alerting, connect a key belonging to a user without those permissions rather than relying on how you word the prompt.

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.