Connect Claude to Pydantic Logfire

Query traces, logs, exceptions and metrics in a Pydantic Logfire project with SQL. Toolspoke puts 7 of its actions behind one MCP endpoint that Claude, Cursor, and Codex all speak.

Connect View pricing

Connection: Provider REST API
Authentication: Read token
Actions exposed: 7
What a call bills: Nothing
Adapter: Maintained by Toolspoke

Connected in three steps

1
Install Pydantic Logfire
Open the marketplace in your workspace, add Pydantic Logfire to the project your agents work in, and it appears on the gateway immediately.
2
Connect the credential
Authenticate with read token. Where to get one, and what it has to be able to reach, is the next section.
3
Point your agent at the gateway
Give your client one address, https://toolspoke.com/mcp. Claude Code takes it as a command, Claude and Claude Desktop add it as a custom connector, and Cursor, Codex and VS Code each read it from a config file of their own.

.mcp.json

{
  "mcpServers": {
    "toolspoke": {
      "type": "http",
      "url": "https://toolspoke.com/mcp"
    }
  }
}

One block covers every tool you have installed. Pydantic Logfire shows up in the client as soon as your policy allows it, and so does everything else you install later.

Where the address goes, per client

Claude Code: Run it in your project, then /mcp to sign in
claude mcp add --transport http toolspoke https://toolspoke.com/mcp
Claude and Claude Desktop: Settings, then Connectors, then Add custom connector
https://toolspoke.com/mcp
Cursor: ~/.cursor/mcp.json, or .cursor/mcp.json for one project
{ "mcpServers": { "toolspoke": { "url": "https://toolspoke.com/mcp" } } }
Codex: ~/.codex/config.toml
[mcp_servers.toolspoke] url = "https://toolspoke.com/mcp"
VS Code: .vscode/mcp.json, or the MCP: Add Server command
{ "servers": { "toolspoke": { "type": "http", "url": "https://toolspoke.com/mcp" } } }

What Pydantic Logfire asks for

Read token. You provide it once, when you install the connector. Toolspoke encrypts it at rest and decrypts it only for the length of a single call, and the gateway attaches it to the outbound request itself, so it is never part of the arguments an agent sends.

Read tokenRequired: Logfire → your project → Settings → Read tokens → Create read token. Copy it at once; it is shown only once. A read token reaches one project.; pylf_v1_us_…
Logfire hostOptional: https://logfire-us.pydantic.dev for the US region, https://logfire-eu.pydantic.dev for the EU region, or the URL of your self-hosted Logfire. It must match the region the token was made in.; https://logfire-us.pydantic.dev

What Claude can do in Pydantic Logfire

7 actions, each one declared and named by the connector rather than discovered at runtime. A workspace policy grants a person all of them, a hand-picked selection, everything on the read side, everything on the write side, or none.

Reads: 7Reads
Writes: 0Writes
Destructive: 0Destructive

Reads

Fetches data and changes nothing.

get_token_info
Show which organization and project this read token reads, its plan and its expiry. Takes no arguments and runs no query, so it uses none of the daily query budget.
query_sql
Run a SQL query (Apache DataFusion dialect, Postgres-like) over the records table (spans and logs: start_timestamp, duration, trace_id, span_id, parent_span_id, span_name, message, level, service_name, is_exception, exception_type, exception_message, attributes) or the metrics table. Always add a LIMIT. The time window defaults to the last 24 hours; widen it with min_timestamp, up to 14 days. Results include log messages and attributes, which can hold your users' data.
list_services
List the services that sent data in the window, with how many records and how many exceptions each one produced. Start here to see what is instrumented.
list_recent_exceptions
The most recent exceptions, newest first, with service, span, exception type and message, and the trace id to pass to query_sql for the full trace.
exception_counts
Count exceptions by service and exception type in the window, most frequent first, with when each was last seen. The quickest answer to "what is breaking".
slowest_spans
The slowest spans in the window, longest first, with service, span name, duration in seconds and trace id.
list_metrics
List the metric names recorded in the window and how many data points each has. Use the names in query_sql against the metrics table.

What it will not do

Enforced by the gateway rather than left to convention, which is why each of these can be stated flatly.

It cannot call anything else: The 7 actions above are the whole of it. A call to any other name is refused before it reaches Pydantic Logfire rather than forwarded on, and connecting your account does not add to the list: it is fixed by the connector, not discovered at run time.
It only reads: Every action here reads. Nothing this connector can do changes anything in Pydantic Logfire.
It reaches no further than your credential: Toolspoke holds no access to Pydantic Logfire of its own. Every call carries the credential you stored and nothing besides, so whatever that credential cannot reach, this connector cannot reach either.
It never hears from Pydantic Logfire: Nothing is pushed to it. There is no webhook, no subscription and no polling, so this connector cannot notice by itself that something changed in Pydantic Logfire. An agent has to ask.
It does not smooth over provider limits: Toolspoke does not retry, queue or back off around Pydantic Logfire's own rate limits. A call that Pydantic Logfire refuses comes back to the agent as a failed call.

Before you connect it

What can Claude do in Pydantic Logfire?

7 named actions: 7 that only read. They include get_token_info, query_sql and list_services. Nothing outside that list is reachable: the connector declares each operation by name rather than proxying whatever an agent asks for.

What credentials does the Pydantic Logfire connector need?

Read token. The connector asks for read token, and optionally logfire host. Values are encrypted at rest and attached to the outbound request by the gateway, so they are never part of the arguments an agent sends and never reach the audit log.

Does the Pydantic Logfire connector work with Cursor and Codex, or only Claude?

Any client that speaks MCP, and every one of them gets the same 7 actions. There is a single address, https://toolspoke.com/mcp. Claude Code adds it with claude mcp add --transport http, Claude and Claude Desktop take it as a custom connector in settings, Cursor reads it from .cursor/mcp.json, Codex from ~/.codex/config.toml, and VS Code from .vscode/mcp.json. Each of them signs in to the gateway itself, so there is no key to paste.

What does the Pydantic Logfire connector not do?

The 7 actions above are the whole of it. A call to any other name is refused before it reaches Pydantic Logfire rather than forwarded on, and connecting your account does not add to the list: it is fixed by the connector, not discovered at run time. Every action here reads. Nothing this connector can do changes anything in Pydantic Logfire. Toolspoke holds no access to Pydantic Logfire of its own. Every call carries the credential you stored and nothing besides, so whatever that credential cannot reach, this connector cannot reach either. Nothing is pushed to it. There is no webhook, no subscription and no polling, so this connector cannot notice by itself that something changed in Pydantic Logfire. An agent has to ask. Toolspoke does not retry, queue or back off around Pydantic Logfire's own rate limits. A call that Pydantic Logfire refuses comes back to the agent as a failed call.

Can I limit which actions an agent can call?

Yes, in two places. The project switches Pydantic Logfire's actions on and off one at a time, for everyone in the project at once, and the screen groups them by read, write and destructive so turning off everything that deletes is one click. An individual agent key can then be narrowed further, to particular toolkits in a project and to particular actions in a toolkit. Whatever it was granted, a key never reaches a project its owner cannot.

What gets recorded when an agent calls Pydantic Logfire?

Every attempt, with the agent that made it and the person that agent belongs to, the full request payload, the response payload, the status, the duration, and what the call cost in sandbox time and model tokens. Values whose key names a secret are masked out before the record is shown to anyone. An operation the connector marks as not retained never has its response body written at all, so the gateway keeps no second copy of what was read.

More observability connectors

Browse all of them