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

Sentry 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 Sentry 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 answers assume a token you created: unquote SENTRY_AUTH_TOKEN, match the region with SENTRY_URL, switch from an auth token to an API key, refresh a stale cache. In a hosted flow there is no token file to unquote and no env to check, which rules out most of that advice. What is left is the grant itself: an expired or revoked OAuth grant produces a 401 on calls while the dashboard still lists the connection as connected, because the connection row and the token's validity are separate facts. The recovery is to reconnect Sentry from the ClawLink dashboard, which re-authorizes and replaces the stored grant. Check your Sentry account's authorized applications first; if the grant was revoked there, the dashboard connection is already stale.

Two causes cover most cases, and neither is a ClawLink bug. The first is scope: the OAuth grant carries the scopes approved at consent, and a tool that needs event:read or project:write will 403 if the grant does not have it; reconnect from the dashboard to approve the full scope set again. The second is org context: the grant is bound to the Sentry org you approved, so a project in a different org returns permission errors rather than an obvious wrong-org message. Sentry's own advice that personal tokens cannot be edited after creation does not apply here, because there is no personal token; when scopes change on your side, the whole grant is re-approved rather than patched.

The answers engines give target coding IDEs: an OAuth handshake in Cursor settings, deleting a cached mcp.json, claude mcp add against https://mcp.sentry.dev/mcp. None of that exists in this setup, which has no local MCP server. Here the cause is lazy schema loading: Sentry tool schemas hydrate on demand the first time a tool runs, so a call in a fresh chat can arrive before the catalog does, and retrying once resolves it. If the same name fails twice, the name is wrong rather than missing, and the error message lists the closest real tools. Reconnect from the dashboard if every Sentry call keeps failing in a fresh chat.

Try reconnecting Sentry 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 Sentry: link your Sentry account once in the browser and Hermes Agent can call the Sentry API through 205 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 Sentry 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 Sentry 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 Sentry working from chat.

No. This connection is a hosted OAuth grant, so there is no SENTRY_AUTH_TOKEN to create in your account settings, export into an env file, or unquote when a guide says so. The manual flow's entire failure surface, which is most of what the search results for "sentry api 401 invalid token" teach, comes from token handling: wrong token type, region mismatch, stale cache, tokens that cannot be edited after creation. None of that exists here. The grant is stored server-side and refreshed for you, and reconnecting from the dashboard is the recovery path when something does break.

The canonical use case is issue triage: sentry_list_an_organizations_issues returns error groups with a default query of unresolved issues at high or medium priority, sentry_get_organization_issue_details accepts both numeric and short IDs, and sentry_bulk_mutate_an_organizations_issues updates status, priority, and assignment across up to 1,000 issues in one call. Around that core sit alerts and monitors, releases and deploys, projects, teams and members, replay sessions, and three Seer tools (sentry_list_seer_ai_models, sentry_start_seer_issue_fix, sentry_retrieve_seer_issue_fix_state) that work on the same issues from chat.

The June 2026 Agentjacking research, which the AI Overview surfaces on this exact question, showed public Sentry DSNs being used to inject prompts into coding agents that ingest error payloads. That attack is about unauthenticated ingestion endpoints, not about an agent holding a scoped grant. This connection is a user OAuth grant, not a public DSN, and the grant's scopes are the visible list from consent: org:read, project:read, event:read, team:read, member:read and the write counterparts, with the 11 total scopes checked against the Composio toolkit on 2026-08-02. The behavioral risk remains real and is not solved by the connection model: an agent reading issue descriptions is reading text written by other people. Keep write actions (status changes, member invites, releases) as things it drafts first and you approve.

No. The connection targets sentry.io. A self-hosted Sentry instance needs its own instance URL, an internal token you manage, and a network path to it, none of which this flow provides.

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.