Talaria

com.newtalariaPublished by newtalaria.comOfficial publisherv1.0.1Updated Oct 29, 2026

Errors, traces, replays and logs from Talaria for coding agents.

VerifiedStreamable HTTPWeb executableDeveloper Tools Security & Monitoring

Server details

Identifier: com.newtalaria/talaria
Transport: Streamable HTTP
Runtime: Web executable
Trust: Verified
Published: Oct 29, 2026

Security note

MCP servers receive tool arguments from a conversation and may perform external actions. Review what a server can do before you enable it, and give it only the access it needs.

The registry verified that this server's namespace belongs to newtalaria.com, which also runs its endpoint. That identifies the publisher; it does not make the server's tools safe.

SourceWeft's checks cover whether the listing is complete and the endpoint answered; they do not review the server's code or what its tools do.

Add a badge to your README

Show readers this MCP server is listed on SourceWeft. The badge links to this page.

[![Listed on SourceWeft](https://sourceweft.com/badge/mcp/com.newtalaria%2Ftalaria.svg)](https://sourceweft.com/mcp/com.newtalaria%2Ftalaria?utm_source=readme&utm_medium=badge)

Overview

AI-generated overview

Hosted Talaria MCP server that lets coding agents investigate production errors, traces, replays and logs, then triage issues and verify fixes.

What you can do: Read project health, traces, spans, queries, sessions and monitors
Investigate a single issue with its pack, trace, replay and suspects, optionally pulling in GitHub context
Verify a fix after a release and optionally update issue status
Manage projects, keys, events, analytics, feature flags and ops tool groups
Read Talaria documentation and connection info through always-on toolsets
When to use it: Useful when a coding agent should work against real production telemetry: diagnosing open errors, following traces and replays, triaging issues, or confirming that a release actually fixed something. It is aimed at teams already sending data to Talaria.
Requirements: Remote Streamable HTTP endpoint at api.newtalaria.com/mcp, with no local package. Sign-in is OAuth 2.1 with PKCE in the browser, or a personal Bearer token from Settings → Integrations. Optional URL paths scope to one organisation or project, and query parameters can restrict to read-only or selected toolsets.

Before you install

OAuth scopes include mcp:read, mcp:write and mcp:keys, and write includes read while keys includes write, so the agent can change data and manage keys. Skills are said to ask before write tools, but that is a convention, not a guarantee. Never use ingest keys (tal_pk_, tal_sk_) or monitor tokens (tal_mon_) as MCP auth; the endpoint rejects them. Use the readonly URL option when only inspection is wanted.

Installation

In SourceWeft

Open Talaria in the dashboard and add it to a workspace.
Connect the required credentials. They are stored privately and never shown on public pages.
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/com.newtalaria%2Ftalaria/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": {
    "talaria": {
      "type": "http",
      "url": "https://api.newtalaria.com/mcp"
    }
  }
}

README

Talaria agent plugin

Hosted Talaria MCP plus skills for coding agents (Copilot, VS Code, Claude Code, Gemini CLI, Cursor, Grok) to investigate production errors, traces, and replays — then triage issues and verify fixes after a release.

MCP: https://api.newtalaria.com/mcp (Streamable HTTP, OAuth 2.1 with PKCE)

No secrets live in this plugin. Agents sign in with OAuth (or a personal Bearer token from Settings → Integrations). Never use a publishable or secret ingest key (tal_pk_…, tal_sk_…, with prod / stg / test / dev in the key) or a monitor token (tal_mon_…) as MCP auth — /mcp rejects them.

Install

Grok Bot (Connect apps)

Open Connect apps in Grok Bot.
Search Talaria (handle talaria) once the marketplace listing is approved — or add a custom MCP server named Talaria at https://api.newtalaria.com/mcp.
Confirm → authorise in the browser → choose your organisation.
Ask something like: "What are the top open errors in production?"

Setup guide: newtalaria.com/mcp

Cursor

Install from cursor.directory/plugins/talaria, or add the MCP server manually by following the Cursor setup guide.

Skills under skills/ teach get-started → triage → verify-fix.

VS Code / GitHub Copilot

Option 1: Install via link (opens VS Code)

Install in VS Code

For VS Code Insiders, use this link.

Option 2: Command line

Option 3: MCP Gallery (when listed)

Search for @mcp talaria in the Extensions view, or run MCP: Add Server → Browse MCP Servers…

GitHub Copilot CLI

Once listed in awesome-copilot:

Or install directly from this repo:

Claude Code

Step 1: Add the self-hosted marketplace

Step 2: Install the plugin

Anthropic directory listing pending.

Gemini CLI

The extension will appear in the gallery at geminicli.com/extensions once the repo has the gemini-cli-extension topic.

Grok Build

Once listed in xai-org/plugin-marketplace, install by plugin name talaria (SHA-pinned remote source). Until then, point Grok Build at this repository's tip.

Skills

Skill	When to use
`talaria-get-started`	Connect, confirm org/project, first smoke read
`talaria-instrument`	Add Talaria SDK, create project/key, send first test event
`talaria-observe-performance`	Read project health, traces, spans, queries, sessions, monitors
`talaria-triage-issue`	Investigate one issue (pack → trace/replay/suspects → optional GitHub)
`talaria-verify-fix`	After a release: `verify_fix` verdict, optional status update

Closed loop: get-started → instrument → observe → triage → verify-fix.

Auth & scopes

Write includes read; keys includes write. Skills ask before any write tool.

Do not paste tal_pk_…, tal_sk_…, or tal_mon_… tokens into MCP clients.

MCP URLs

URL	Behaviour
`https://api.newtalaria.com/mcp`	All orgs you belong to
`https://api.newtalaria.com/mcp/<org>`	One organisation
`https://api.newtalaria.com/mcp/<org>/<project>`	One project
`…?readonly=true`	Read tools only
`…?toolsets=issues,performance`	Limit tool groups

Always-on toolsets: docs, connection. Others: projects, keys, issues, events, performance, analytics, flags, ops.

Repository layout

plugin.json                  # Agent Plugins 1.0 manifest (VS Code, Copilot CLI, Cursor)mcp.json                     # Agent Plugins MCP wiring (type: streamable-http).cursor-plugin/plugin.json   # Cursor plugin manifest.grok-plugin/plugin.json     # Grok Build style manifest.claude-plugin/              # Claude Code plugin + self-hosted marketplace.mcp.json                    # MCP wiring for Claude Code / Grok Build (type: http)gemini-extension.json        # Gemini CLI extension (httpUrl)server.json                  # Official MCP Registry entryskills/                      # Agent skills (closed loop)scripts/                     # mcp-tools.json allowlist + check-skill-tools.mjsassets/logo.png              # 512×512 mark on solid plateLICENSE                      # MIT · Talaria Labs

plugin.json                  # Agent Plugins 1.0 manifest (VS Code, Copilot CLI, Cursor)mcp.json                     # Agent Plugins MCP wiring (type: streamable-http).cursor-plugin/plugin.json   # Cursor plugin manifest.grok-plugin/plugin.json     # Grok Build style manifest.claude-plugin/              # Claude Code plugin + self-hosted marketplace.mcp.json                    # MCP wiring for Claude Code / Grok Build (type: http)gemini-extension.json        # Gemini CLI extension (httpUrl)server.json                  # Official MCP Registry entryskills/                      # Agent skills (closed loop)scripts/                     # mcp-tools.json allowlist + check-skill-tools.mjsassets/logo.png              # 512×512 mark on solid plateLICENSE                      # MIT · Talaria Labs

Validate skill tool names against scripts/mcp-tools.json:

License

Source: README.md at commit 1a35fcc

Tools

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

v1.0.1LatestOct 29, 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

Method	Notes
OAuth (preferred)	Browser sign-in; scopes `mcp:read` / `mcp:write` / `mcp:keys`
Personal token	Settings → Integrations; send as `Authorization: Bearer …`