Sentry vs Mushi Mushi
Sentry is built around what the code threw, with a User Feedback widget and Session Replay alongside. Mushi starts from what the user reported, ingests Sentry’s errors too, explains each one in plain English, and hands your agent a fix prompt to start from. One queue, with or without Sentry.
That is the whole difference in one paragraph. The rest of this page is the detail: what each product captures, what the free tiers include, what the AI layer costs, and how they run side by side.
Facts checked against the linked sources. Next review due . Rows marked “unverified — check before publishing” could not be confirmed on that date.
What each one is
Sentry: Error and performance monitoring: SDKs capture exceptions, traces and releases, and Seer is a paid AI debugging agent on top.
| Claim | Sentry | Source |
|---|---|---|
| Free tier | Developer plan: $0, 5k errors, limited to one user | checked 2026-09-21 |
| Paid tiers | Team $26/mo · Business $80/mo · Enterprise custom | checked 2026-09-21 |
| Seer (AI debugging agent) | An add-on to the Team, Business or Enterprise plan at $40 per active contributor per month, billed as its own monthly charge, separate from pay-as-you-go. An active contributor is anyone who opens two or more pull requests in a month to a Seer-enabled repo. The older $20-a-month-plus-credits pricing stopped being offered in January 2026. | checked 2026-09-21 |
| Self-host | Yes: getsentry/self-hosted (Docker Compose; 4 cores, 16 GB RAM + 16 GB swap, 20 GB disk). Seer is closed source and not available in self-hosted Sentry. | checked 2026-09-21 |
Mushi Mushi: An in-app bug reporter for production apps: a user reports from the widget, Mushi attaches the screenshot and console/network tail, writes a plain-English diagnosis, and hands a fix prompt to Cursor or Claude Code over MCP.
| Claim | Mushi Mushi | Source |
|---|---|---|
| Free tier | Free Cloud: 50 diagnoses per month (hard stop; later reports are kept, not diagnosed), 1 seat, 7-day retention, no card | checked 2026-09-21 |
| Paid tiers | Indie $15/mo (500 diagnoses) · Pro $49/mo (2,000 diagnoses) | checked 2026-09-21 |
| License | SDKs MIT; server AGPLv3 | checked 2026-09-21 |
| Self-host | Free, whole stack, one command onto your own Supabase project (free tier works); bring your own Anthropic key (SELF_HOSTED.md) | checked 2026-09-21 |
| Editor loop | MCP server for Cursor / Claude Code; no second LLM key in the MCP process | checked 2026-09-21 |
| Diagnosis accuracy | An LLM reading of the report and your repo. It can be wrong or name the wrong file; read it before you apply the fix. | checked 2026-09-21 |
Side by side
| Question | Sentry | Mushi |
|---|---|---|
| What it sees | Errors and performance, plus Session Replay and a User Feedback widget | User reports from the widget with a screenshot, plus the errors Sentry catches, in one queue |
| What lands in your queue | A stack trace | A short user note plus the screenshot they were looking at |
| Repeat bugs | Repeat events of the same error group into one issue | Repeat user reports of the same broken button collapse to one row, even when nothing threw |
| What you learn from fixes | A resolved issue reopens as a regression if it recurs; there is no lessons file for your editor | Past fixes become rules your editor sees on the next PR (.mushi/lessons.json) |
| Closing the loop | Resolve by hand, from a commit that references the issue, or in a release; Seer, the paid AI debugging add-on, is on Team and above | A draft PR from your agent; merging it resolves the linked Linear issue, and the linked Sentry issue once the Sentry plugin is on (Indie and above, or self-hosted) |
| Reporter attribution | The user you set with setUser, or the name and email on a feedback submission; no credit when it is fixed | "Fixed by Kenji" in the changelog and an SDK toast, once Releases is enabled |
| From your IDE | Sentry's MCP server lets Cursor or Claude Code read issues and ask Seer | Cursor reads the report + relevant lessons and proposes the diff |
| Where it runs | Their cloud, or yours with getsentry/self-hosted (no Seer) | Yours, ours, or both |
| Free tier | Developer: 5k errors/mo, one user | Free Cloud: 50 diagnoses/mo, 1 seat, 7-day retention, no card |
| AI diagnosis | Seer: $40 per active contributor per month, an add-on to Team ($26/mo), Business or Enterprise | Included in every diagnosis, free tier onward |
| Self-host | getsentry/self-hosted (no Seer); 4 cores / 16 GB RAM minimum | AGPLv3 server on your own Supabase project, one command; diagnosis with your own LLM key |
Where they overlap, and where they do not
Both capture thrown exceptions from the browser and from Node. Sentry goes much deeper on the code side: performance tracing, release health, profiling, source-map symbolication across many platforms, and a mature alerting model. If your app is instrumented to the hilt and a team reads the dashboards, Sentry alone is a fine choice.
Mushi starts from the other end. The signal is a person saying “this is broken” inside your app, and the output is a diagnosis you can read without opening a stack trace, plus a fix prompt scoped to the files involved. It matters most when you did not write most of the code, which is the normal case for an app built with Cursor, Claude Code or Lovable.
Running Mushi alongside Sentry
Keep Sentry where it is. Point a Sentry issue-alert webhook at Mushi and every error lands in the same queue as user reports, deduped and diagnosed; with the outbound Sentry plugin (Indie and above, or self-hosted), Mushi resolves the linked Sentry issue when the fix merges. The setup is three steps on the Sentry plugin page, and there is a migration guide for the enrich-or-standalone decision.
Three things to know before you compare bills. Seer is an add-on to the Team, Business or Enterprise plan, priced per active contributor and billed as its own monthly charge, separate from pay-as-you-go; the price and what counts as an active contributor are in the Sentry table above, from Sentry’s pricing docs . Seer is not available in Sentry self-hosted, so the AI layer is cloud-only on their side. And Mushi’s diagnosis runs on every plan, including self-hosted with your own LLM key.
Pick Sentry if
- You need tracing, profiling, release health or platform coverage Mushi does not have (native iOS symbolication, for example).
- A team already reads Sentry every day and the workflow works.
Pick Mushi if
- Users tell you about bugs and you want the report to arrive already explained, with the screenshot they were looking at.
- You fix bugs in Cursor or Claude Code and want the diagnosis, fix context and past lessons in the editor over MCP, without a second LLM key.
- You want the AI layer on the free tier or on your own server.
Start free — 50 diagnoses/mo, no card
Or run npx mushi-mushi in your app and send a test report from the console.
Self-hosting is one command.
Frequently asked questions
- Is Mushi a replacement for Sentry?
- It can be, or it can run alongside Sentry. Sentry is built around what the code threw, with a User Feedback widget and Session Replay alongside. Mushi starts from what the user reported, ingests Sentry's errors too, explains each one in plain English, and hands your agent a fix prompt to start from. One queue, with or without Sentry.
- Can I run Mushi alongside Sentry?
- Yes. Point a Sentry issue-alert webhook at Mushi and errors land in the same queue as user reports, deduped and diagnosed; that inbound path works on every plan. Resolving the Sentry issue when the Mushi fix merges needs the outbound Sentry plugin, which is on Indie and above or self-hosted. Setup is on the Sentry plugin page.
- Does Sentry have an AI fix loop?
- Seer is Sentry's AI debugging agent. Sentry sells it as an add-on to the Team, Business or Enterprise plan at $40 per active contributor per month, where an active contributor is anyone who opens two or more pull requests in a month to a Seer-enabled repo. It is billed as its own monthly charge, separate from pay-as-you-go, and it is not available in self-hosted Sentry.
- What does Mushi cost?
- Free Cloud is 50 diagnoses a month with no card. Indie is $15 a month, Pro is $49 a month. Self-hosting the AGPLv3 server is free.
Related: Sentry alternative for AI-built apps · Sentry alternatives for solo founders · Sentry Replay coexistence