Skip to Content
vnext · upcoming Unreleased Read the changelog →
CompareJam vs Mushi

Jam vs Mushi Mushi

Jam and Mushi both put console and network logs next to a bug so nobody has to ask “can you reproduce that?”. The difference is who presses the button. With Jam, a person on or near your team records the bug in Chrome. With Mushi, the user who hit the bug reports it from inside your production app, and the report arrives with a diagnosis already written.

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

Jam: A Chrome extension a person uses to record a bug: video plus console, network, device metadata and user actions, sent to Jira or Linear. Recording Links let people outside your workspace record too.

ClaimJamSource
Free tier$0: 30 Jams/mo, recordings up to 5 min, 5 Recording Links/mo, 5 creator seats, MCP, webhooks, Jira/Linearchecked 2026-09-21
Team tier$14 per creator per month billed yearly: unlimited Jams, 15-min recordings, 150 Recording Links/mo, 200 AI summarieschecked 2026-09-21
EnterpriseCustom, yearly billing only: SAML/SSO, audit logs, automatic data deletionchecked 2026-09-21
Agent access"Debug Jams via MCP" is on the free tierchecked 2026-09-21
Self-hostNot offered on the pricing pagechecked 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.

ClaimMushi MushiSource
Free tierFree Cloud: 50 diagnoses per month (hard stop; later reports are kept, not diagnosed), 1 seat, 7-day retention, no cardchecked 2026-09-21
Paid tiersIndie $15/mo (500 diagnoses) · Pro $49/mo (2,000 diagnoses)checked 2026-09-21
LicenseSDKs MIT; server AGPLv3checked 2026-09-21
Self-hostFree, 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 loopMCP server for Cursor / Claude Code; no second LLM key in the MCP processchecked 2026-09-21
Diagnosis accuracyAn 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

QuestionJamMushi
Who presses recordA teammate, tester or support agent with the extension, or an outsider via a Recording LinkYour end user, from the widget or a phone shake, inside the production app
What arrivesA video with console, network, device metadata and user actionsThe user's sentence, a screenshot, the console/network tail, route and timeline
What you read firstThe recordingA plain-English diagnosis: what broke, why, which files
Repeat reportsOne Jam per recordingSame bug collapses to one row
Agent hand-offMCP exposes the Jam's logs to your agentMCP exposes the diagnosis, fix context and lessons; optional agent-drafted PR
Free tier30 Jams/mo, 5-min recordings, 5 creator seats50 diagnoses/mo, 1 seat, 7-day retention, no card
Paid entry$14 per creator per month, billed yearly$15/mo Indie (500 diagnoses)
Self-hostNot offeredFree, AGPLv3

Where they overlap

Both attach the technical context a developer needs: console output, network requests, device details and the sequence of user actions. Both expose that context to a coding agent over MCP, and Jam’s “Debug Jams via MCP” is on the free tier, which is a fair point in its favour for a team that already uses Jam for QA.

Where they do not

Jam is a recording tool. The unit is a Jam: a video a human made, sent to a tracker such as Jira or Linear. Reading the video is still your job, and Recording Links are the way to get one from someone outside the workspace.

Mushi is a reporting SDK plus a diagnosis. The unit is a report: the user’s sentence, a screenshot, the console/network tail, the route and a timeline of what they did before. Mushi reads that against your repo and writes what broke and why. Repeat reports collapse into one row, and each fix becomes a lesson your editor sees next time.

Using both

Sensible, and common: Jam for internal QA passes and support-desk recordings, Mushi for reports from real users in production and for the fix loop in Cursor or Claude Code. There is no direct integration between the two today; if you want a Jam link on a Mushi report, put it in the report’s metadata.

Pick Jam if

  • Your bugs are found by teammates, testers or support agents who can run a Chrome extension.
  • A video is the artefact your tracker workflow expects.

Pick Mushi if

  • Your bugs are found by end users on devices you do not control, including mobile (React Native, Capacitor, Flutter, iOS, Android).
  • You want the first thing you read to be the diagnosis, not the recording.
  • You want the report and fix context inside Cursor or Claude Code.

Start free — 50 diagnoses/mo, no card

Frequently asked questions

Is Jam a bug reporter for end users?
Mostly for people on or near the team. Jam is a Chrome extension a teammate, tester or support agent uses to record a bug; Recording Links let someone outside the workspace record too. Mushi is an SDK inside your production app, so the person who hit the bug reports it without installing anything.
Can I use Jam and Mushi together?
Yes, and it is a sensible split: Jam for internal QA and support recordings, Mushi for reports from real users in production and for the diagnosis-to-fix loop in your editor. There is no direct integration between the two today.
Which one gives my coding agent more?
Both have MCP. Jam's MCP hands the recording's console and network logs to your agent. Mushi's MCP hands over the plain-English diagnosis, the fix context scoped to files, past lessons, and can dispatch a draft PR.
What do the free tiers cover?
Jam Free (2026-09-21): 30 Jams a month, recordings up to 5 minutes, 5 Recording Links, 5 creator seats. Mushi Free Cloud: 50 diagnoses a month, no card, unlimited reports.

Related: Sentry vs Mushi · Debug apps built with Cursor · MCP quickstart

Last updated on