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

Discord 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 Discord 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.

The standard fixes for this error assume a bot: check the bot's permissions, role hierarchy, gateway intents, or OAuth2 scopes, then re-invite the bot. None apply to a hosted user connection, because there is no bot to re-invite. Two causes cover it. The connected account may lack permission for the action in that guild: the agent acts as you, so a guild where you are not a member or lack the role fails the same way you would fail. Or the agent sent a placeholder id instead of a real one. Run discord_list_my_guilds and discord_get_my_guild_member to confirm what the account can do, then retry with a real id.

Ask the agent to diagnose it:

Run discord_get_my_user and discord_list_my_guilds, tell me which account is connected and which guild the failing call targets. Do not retry until both have returned.

This error comes from the self-hosted world: copying the client secret instead of the bot token, a typo in the env file, or a leaked token that Discord auto-revoked. None of that exists here, because there is no bot token in the flow. If the connection itself fails, it fails on Discord's side of OAuth, and the fix is to reconnect Discord from the ClawLink dashboard and complete the consent screen again.

Ask the agent to diagnose it:

Tell me the exact error text of the failed Discord call. If it names a token, reconnect the Discord connection from the dashboard.

Discord rate limits are per resource, and this connection runs under the Discord application ClawLink operates rather than a bot token you manage. The response carries a retry-after value, which is the interval to respect; pacing bursty calls prevents most 429s. No specific limit is claimed here, because the limits belong to the underlying application's tier and are not published on this page.

Ask the agent to diagnose it:

Report the exact status and any retry-after value from the failed Discord call, and do not retry until it has passed.

The answers for this phrase live in Claude Code and Cursor forums: absolute paths in claude_desktop_config.json, a missing DISCORD_TOKEN env var, a cache clear, or a /mcp Reconnect. There is no local MCP config in a hosted connection. A tool-not-found response here means the tool name was wrong or the schema catalog had not loaded; schemas load on demand, so one retry clears the second case, and the error message names the closest real tools for the first.

Ask the agent to diagnose it:

List the Discord tools you have access to and use the exact name for what I asked.

Try reconnecting Discord 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 Discord: link your Discord account once in the browser and Hermes Agent can call the Discord API through 27 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 Discord 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 Discord 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 Discord working from chat.

No. The AI Overview's "you will always need a free Discord Bot Token" is true for a bot you run yourself; this connection is user-account OAuth, so no bot exists in the flow. You approve the connection on Discord's consent screen, the scopes (identify, email, connections, guilds, guilds.join, guilds.members.read, messages.read, openid) are listed before you click allow, and the agent calls the tools as your account. There is no token to copy, store, or rotate.

No, and this is a real limit worth naming. The 27 tools act on your user account: profile, guild membership, role connections, entitlements, and invites. There is no send-message tool in the set, because the connection has no bot identity and no message-send scope. If you need bot-style posting to channels, that is the separate Discord Bot integration (hermes/discord-bot), which carries bot permissions. For this user-account integration, the agent reads servers, resolves invites, and manages your profile and memberships.

The common framing is bot-token hygiene: keep the token secret, scope it tightly, avoid over-privileged bots. That framing does not transfer to this connection, because there is no token to leak. The grant is a user OAuth authorization, visible under your account's authorized applications in Discord, and revoking it there or in the ClawLink dashboard ends access immediately. What transfers is the behavioural risk: the agent reads messages and server content written by other people, and that content can carry prompt-injection attempts. Keep write-capable tools confirm-first and treat anything the agent reads in a guild as untrusted input.

Several tools in this integration say it in their own schema: discord_edit_application_command_permissions, discord_get_application_command_permissions, and discord_get_batch_application_command_permissions each document that they require an OAuth2 bearer token and that bot tokens will error. That is by design. The connection is your user account, and those permission tools act on your application's commands. A self-hosted bot setup expecting those calls to work with a bot token is using a different integration model.

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.