BugMojo

io.github.viveksinraPublished by GitHub account viveksinrav0.2.1Updated Oct 31, 2026

Bug tracking for AI agents: session replay, console/network logs, repro packs, test plans, evidence.

VerifiedStreamable HTTPWeb executableProductivity & WorkflowDeveloper Tools

Overview

AI-generated overview

Connects AI coding agents to BugMojo bug tracking so they can list, search, read, create and update bugs with captured session replay and log context.

What you can do
  • List and filter bugs by project, status, priority or assignee, and search bug titles and descriptions
  • Read a single bug's full details, including description, comments, labels, recordings and screenshots
  • Create new bugs and update status, priority, assignee, title or description
  • Add comments to existing bugs
  • The hosted OAuth endpoint additionally covers regression suites and runs, QA workflow, agents, session-replay summaries and inline console/network logs
When to use it
Useful when a team tracks bugs in BugMojo and wants a coding agent to triage, summarize or file bugs with the full browser capture context instead of a one-line description.
Requirements
Either the remote HTTP endpoint at bugmojo.com/api/mcp with OAuth browser login, or the npm package run via npx with Node.js 18 or newer. The npm route needs a BugMojo API key in BUGMOJO_API_KEY (starts with bm_key_, generated under Settings > API Keys); BUGMOJO_API_URL is optional and only for self-hosted or local instances. Network access to the BugMojo instance.
Before you install
The API key is a secret passed in the environment; a write-scoped key is needed for create_bug, update_bug and add_comment, which change bug data. Bug content, comments and captured logs or recordings are sent to the configured BugMojo instance, so check that self-hosted URLs point where you expect. The optional mode=admin query parameter exposes admin-management tools.

Installation

In SourceWeft

  1. Open BugMojo in the dashboard and add it to a workspace.
  2. Connect the required credentials. They are stored privately and never shown on public pages.
  3. Enable the server for the chats that should use its tools.

Web executable via Streamable HTTP. Remote servers run from the web runtime once configured in a workspace.

Use with an agent

Paste this prompt into Claude Code, Codex, Cursor or another agent — or into SourceWeft chat. The agent reads this server's install guide, shows you its publisher, endpoint and sign-in, and adds it to your client once you agree. Keys never go into a command.

Read https://sourceweft.com/mcp/io.github.viveksinra%2Fbugmojo/install.md and connect the MCP server it describes to my client. Before changing any config, show me its publisher, endpoint and sign-in method and wait for my OK. Never put an API key in a command or in a file I would commit; ask me where it should come from.

Read the install guide the agent follows

Other MCP clients

Add this to your client's mcpServers config. The client completes OAuth sign-in on first connect.

{
  "mcpServers": {
    "bugmojo": {
      "type": "http",
      "url": "https://www.bugmojo.com/api/mcp"
    }
  }
}

README

@bugmojo/mcp-server

MCP server that connects AI coding agents — Claude Code, Cursor, VS Code, Claude Desktop — to BugMojo bug tracking.

[npm version] [license: MIT]

Why

BugMojo captures pixel-perfect bug reports straight from the browser — session replay, console logs, network requests, screenshots. This MCP server hands that context to your AI coding agent: it can list, search, and read bugs, file new ones, update status/priority/assignee, and add comments over BugMojo's API-key-authenticated REST API. Your agent fixes bugs with the full capture context instead of a one-line description.

Want the full surface? Regression suites/runs, testing workflow, agents, session-replay summaries, and inline console/network logs are available on the hosted, OAuth-authenticated HTTP endpoint at https://www.bugmojo.com/api/mcp — see Full surface (OAuth HTTP) below. This npm package intentionally stays a minimal bugs shim so every tool maps to a real endpoint (nothing 404s).

Install

Requires Node.js 18+. No install needed — MCP clients run it with npx:

bash
npx -y @bugmojo/mcp-server

Or install the bugmojo-mcp command globally:

bash
npm install -g @bugmojo/mcp-server# orpnpm add -g @bugmojo/mcp-server# oryarn global add @bugmojo/mcp-server

It speaks MCP over stdio and reads two environment variables (below).

Quickstart (Claude Code, 60 seconds)

  1. Generate an API key in your BugMojo dashboard under Settings → API Keys (it starts with bm_key_).
  2. Register the server:
bash
claude mcp add bugmojo -e BUGMOJO_API_KEY=bm_key_your_key_here -- npx -y @bugmojo/mcp-server
  1. Ask Claude: "List the open CRITICAL bugs and summarise the newest one."

Configuration

VariableRequiredDescription
BUGMOJO_API_KEY✅Your BugMojo API key (starts with bm_key_). Generate one in the dashboard under Settings → API Keys.
BUGMOJO_API_URL—Base URL of your BugMojo instance. Defaults to https://www.bugmojo.com — only set this for self-hosted or local instances (e.g. http://localhost:3000).

Client setup

Claude Code

Either use the CLI (per-user):

bash
claude mcp add bugmojo -e BUGMOJO_API_KEY=bm_key_your_key_here -- npx -y @bugmojo/mcp-server

Or check a project-scoped .mcp.json into your repo root so the whole team gets it:

json
{  "mcpServers": {    "bugmojo": {      "command": "npx",      "args": ["-y", "@bugmojo/mcp-server"],      "env": {        "BUGMOJO_API_KEY": "bm_key_your_key_here"      }    }  }}

Claude Desktop

Add to claude_desktop_config.json (Settings → Developer → Edit Config):

json
{  "mcpServers": {    "bugmojo": {      "command": "npx",      "args": ["-y", "@bugmojo/mcp-server"],      "env": {        "BUGMOJO_API_KEY": "bm_key_your_key_here"      }    }  }}

Cursor

~/.cursor/mcp.json (global) or .cursor/mcp.json (per project):

json
{  "mcpServers": {    "bugmojo": {      "command": "npx",      "args": ["-y", "@bugmojo/mcp-server"],      "env": {        "BUGMOJO_API_KEY": "bm_key_your_key_here"      }    }  }}

VS Code (MCP)

.vscode/mcp.json:

json
{  "servers": {    "bugmojo": {      "command": "npx",      "args": ["-y", "@bugmojo/mcp-server"],      "env": {        "BUGMOJO_API_KEY": "bm_key_your_key_here"      }    }  }}

Any other MCP client

Point it at the same stdio command: npx -y @bugmojo/mcp-server (or bugmojo-mcp if installed globally) with BUGMOJO_API_KEY in the environment. Add BUGMOJO_API_URL only if you self-host.

Tools

ToolWhat it does
list_bugsList/filter bugs by project, status, priority, assignee (paginated).
get_bugFull details of one bug: description, comments, labels, recordings/screenshots.
search_bugsFull-text search across bug titles and descriptions.
create_bugCreate a bug in a project (title, description, priority, assignee). Needs a write-scoped key.
update_bugUpdate status, priority, assignee, title, or description. Needs a write-scoped key.
add_commentAdd a comment to a bug. Needs a write-scoped key.

That's the whole npm package — six tools, each mapping to a real /api/v1/bugs/* endpoint.

Full surface (OAuth HTTP)

Regression suites/runs, the testing (QA) workflow, agents, session-replay summaries (summarize_replay), and inline console/network logs (get_bug_console_logs, get_bug_network_logs) are served by the hosted, OAuth 2.1-authenticated streaming endpoint — not this npm package. Point an HTTP-capable MCP client at:

https://www.bugmojo.com/api/mcp

The client opens a browser login the first time (no API key to manage). Optional query params: ?workspace=<slug> binds the connection to a workspace, and ?mode=admin additionally exposes the admin-management tools.

jsonc
// Claude Code / clients that support remote MCP servers{  "mcpServers": {    "bugmojo": {      "type": "http",      "url": "https://www.bugmojo.com/api/mcp"    }  }}

Example

Once configured, ask your agent things like:

"List the open CRITICAL bugs in project X, open the newest one, and summarise the repro steps."

The agent calls list_bugs (status=OPEN, priority=CRITICAL), then get_bug for the full context — including the session replay BugMojo captured in the browser.

How it works

This package is a thin, stateless client: it translates MCP tool calls into authenticated HTTPS requests to the BugMojo REST API (/api/v1). All business logic and data live in your BugMojo instance — nothing is stored locally.

Troubleshooting

"BUGMOJO_API_KEY environment variable is required" The key isn't reaching the server process. Put it in the env block of your MCP config (or pass -e BUGMOJO_API_KEY=... to claude mcp add) — shell exports are not inherited by every MCP client.

"BUGMOJO_API_KEY must start with 'bm_key_' prefix" You copied the wrong value (e.g. a session token or a truncated key). Regenerate the key in the dashboard under Settings → API Keys and copy the whole bm_key_... string.

"Authentication failed ... Check your BUGMOJO_API_KEY" (401) The key was revoked or belongs to a different instance. Generate a fresh key; if you self-host, also confirm BUGMOJO_API_URL points at that instance.

"Permission denied" (403) on create/update/comment Your key is read-only. create_bug, update_bug, and add_comment need a write-scoped key.

"fetch failed" / ECONNREFUSED The server can't reach BUGMOJO_API_URL. The default is https://www.bugmojo.com; if you set a self-hosted/local URL, make sure that instance is running and reachable.

Server disconnects immediately on first run Make sure your config uses "args": ["-y", "@bugmojo/mcp-server"] — without -y, npx's first-run "Ok to proceed?" prompt hangs the stdio handshake. Also confirm node --version is 18 or newer.

Related packages

PackageUse it for
@bugmojo/widgetFramework-agnostic on-site feedback + capture widget core
@bugmojo/reactReact / Next.js SDK — provider, error boundary, hooks
@bugmojo/react-nativeReact Native + Expo SDK — shake-to-report
@bugmojo/cliPull a bug's Playwright repro pack, verify fixes locally

Links

License

MIT © Softech Infra

Source: packages/mcp-server/README.md at commit 42e139d

Tools

0
This server asks you to sign in, so SourceWeft could not read its tools without an account and has not indexed them. What a connected account sees can depend on that account and its permissions.

Version history

1
  1. v0.2.1LatestOct 17, 2026

Score

85% of the applicable checks met

How complete this listing is, computed from the registry entry, the endpoint and public sources. Not a rating of the server and not a statement that it is safe or working today: reachability is a dated check, and no installs or reviews go into the score.

  • Reached in the last 30 daysRequired25 pts
  • Auth declaredRequired10 pts
  • Tools listed15 pts

    Answer tools/list without a credential, or ship a package that starts in a sandbox without one.

  • DescriptionRequired15 pts
  • Logo10 pts
  • Documentation15 pts
  • License10 pts