Think inside your AI world.
How do I connect Sentry to Claude?
Sentry over MCP puts the error stream where the debugging conversation already is — projects, issue detail, retrieval, and Seer's fix flows.
Direct answer: add Sentry's MCP server in Claude via Settings → Connectors and authorise your org ('claude mcp add' in Claude Code). Claude reads projects, issue details and errors, and can work with Seer's fix flows. The tool serving 150k+ organisations and four million developers becomes a conversational surface: which issues spiked, what the stack trace says, what changed around the regression.
Triage in the conversation
What is new since the deploy, which issue affects the most users, what the breadcrumbs show before the crash — read live during the debugging session that needs them, instead of in a second window feeding a first.
The budget the error stream doesn't know it has
Worked case: you ratified an error budget — more than five new unique crashes in the twenty-four hours after a release triggers a ship-freeze until triage clears, set after a release train once outran its own wreckage. Sentry counts crashes precisely; the freeze threshold is a promise about the counts, held by the team, not the tool.
The stream, read against the budget
Through Unl the post-release read arrives adjudicated: seven new uniques in twenty-two hours — freeze, by the rule the release train wrote. 'Is it safe to keep shipping?' becomes a question your own policy answers on live data, with the wreckage story attached for whoever asks why the freeze exists.
Sentry to Claude reads the error stream; Sentry through Unl reads it against the error budget you ratified — the freeze fires as your decision, not as somebody's judgement call at 6pm.
Reads through Unl arrive with measured context — in the presence of the decisions you’ve already settled. The reach lane is live: one box, paste anything. If it speaks MCP, Unl can reach it. Readings arrive unprompted, the data beside the criterion; Unl is a courier, not a warehouse, and keeps only your keys and the frame.
Read further
Questions people ask
What does Claude read from Sentry?
Projects, issue details and error retrieval, plus Seer fix flows — the observational and remediation surface of the error tracker.
Does the connection work for other MCP clients?
Yes — the server is protocol-standard; ChatGPT and other MCP clients connect identically.
What makes an error budget worth ratifying rather than documenting?
Documentation gets consulted after the argument starts; a ratified budget arrives inside the read, already applied. The freeze is declared by the rule you set — with its reasoning — rather than proposed by whoever noticed.
What this is
Think inside your AI world — you stay in command
Unlimitless (Unl to friends) holds what you've settled, reads what your tools are showing, and catches what's changed out in the world — and hands your AI whatever bears on the work, the moment it's needed, without you asking. The right thing, in front of the model, unprompted, with you in command of the call. So you keep moving toward what you set out to build, on top of everything you've already decided.
It plugs into Claude, Claude Code, ChatGPT and Cursor as an MCP connector. Quick to connect, in a couple of steps.
Unlimitless is open now to invited Alpha. Apply for the Beta waitlist to come in ahead of the full launch:
Alpha is invite-only · free at launch.