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

Apollo 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 Apollo 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.

This is the card ChatGPT quotes back, and the causes are the two it lifts. First, a placeholder argument: an agent that has not listed anything guesses an id, and strings like YOUR_COMPANY_ID are structurally valid, so the call reaches Apollo and returns as a permissions failure rather than a bad-argument failure. List first, act second: run apollo_people_search or apollo_search_contacts and use an id from the response. Second, genuine account scope: what the connected Apollo user can see is decided by that account's plan, seats, and team membership, and the connection cannot see past them. apollo_get_auth_status reports which user and team the connection acts as, which separates a wrong-id problem from a real permissions problem in one call.

Ask the agent to diagnose it:

Run apollo_get_auth_status and tell me which Apollo user and team this connection acts as. Then run one search tool and show me the results. Do not retry the failing call until both have returned.

Enrichment consumes credits and Apollo rate-limits it, with a 429 and a fixed-window reset when the limit is hit. An agent walking a list and enriching every row is exactly the traffic shape that meets the limit. The fix is to change the shape of the work: search first, enrich only the rows that matter, and if a bulk enrichment is genuinely needed, run it in stages across turns rather than one sweep. The agent should report the 429 rather than retry blindly, because the window resets on Apollo's schedule, not yours.

Ask the agent to diagnose it:

Tell me how many enrichment calls you have made and which tool returned 429. Quote the exact error and any reset window it names. Do not call Apollo enrichment again until I reply.

The search results answer this with Apollo GraphQL's apollo-mcp-server: .graphql file paths, npx config, IDE restarts. Wrong product entirely. Here the failure is the hosted-catalog pattern: schemas load on demand, so the first Apollo call in a fresh session can arrive before the catalog and needs one retry. If no Apollo tools appear at all, the connection or the pairing is incomplete, so confirm the plugin was installed with --enable, that both hermes clawlink begin and hermes clawlink finish ran, and that Apollo shows as connected in the dashboard. If one name fails twice, it is wrong rather than missing, and the error lists the nearest real tools.

Ask the agent to diagnose it:

List the Apollo tools you actually have available and tell me which one searches people. Then use that exact name.

Double-check that the API key for Apollo 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 Apollo: link your Apollo account once in the browser and Hermes Agent can call the Apollo API through 48 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 Apollo in the dashboard and Hermes can use it from the next message — no config files, and the Apollo key you paste is stored server-side instead of in your environment.

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

Two different companies' products. Apollo.io is the sales data platform: contacts, deals, enrichment, sequences. Apollo GraphQL is the open-source GraphQL server, whose apollo-mcp-server and .graphql config dominate search results for "Apollo MCP". Everything on this page is about Apollo.io. If an answer mentions schema files and npx, it is about the GraphQL server, not this integration.

Fair question, because it exists: mcp.apollo.io is hosted, uses OAuth 2.0, needs no API key, and its docs were updated 2026-07-30, which is why the engines now recommend it. What it changes is the setup difference, not the features. Their endpoint is another MCP server you point Hermes at: an entry in ~/.hermes/config.yaml plus an OAuth flow you run. This route has no config file at all: the plugin install and one pairing handshake are the entire client-side story, the key is pasted once for 48 tools, and Apollo sits alongside your other connected accounts with one place to revoke everything. If you are comfortable maintaining a config file, their server is real; if you want zero setup, this is the one.

Search and enrichment are different tools with different economics. The search tools, apollo_people_search and apollo_organization_search, return records that match filters. Enrichment, apollo_people_enrichment and the bulk variants, is the call that pulls the fuller record and consumes credits, and it is where Apollo rate-limits bite: the taught free-tier figures are 50 enrichment requests per minute, 200 per hour, and 600 per day, with a 429 and a fixed-window reset when you exceed them. A search that returns contact records without emails is normal; asking the agent to enrich instead of search is the credit decision to make deliberately.

Apollo holds contact data, so the answer is the same as for any CRM: keep the credential out of the agent's hands. Here the key is stored server-side, the agent never receives it, and the connection is revocable from the dashboard in one click. What the agent can see is what your Apollo account can see; the guidance about not handing master keys to third-party builders is exactly the risk this setup removes.

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.