TAI42 — the OS for AI tools

Collect metrics and traces from a running server. The runtime records what runs. Tool and agent runs produce traces; aggregate counts and latencies roll up into metrics. You query both from the CLI. A monitoring backend plugin persists and serves the data; the monitor extension traces a standalone tool call as one span.

Query metrics

Query aggregate metrics over a time range. --from and --to take an ISO instant or a relative token (7d, 30d); --granularity is hour, day, or week.

tai obs metrics --from 7d --granularity day

List and read traces

List runs

List observability runs, or download the filtered list with --export.

tai traces list

Read one trace

Get one run’s full trace, or download it as JSON with --export.

tai traces get <run-id>

Trace a standalone tool call

A tool call outside a flow or agent has no trace of its own. Load the monitor extension and stack it onto a tool to record the call as one live span. Load the module, then attach monitor to a tool.

manifest.yml

extensions_modules:
  - tai42_skeleton.extensions.builtin.monitor

tools:
  - title: my_tools
    module: myapp.tools
    include: [greet]
    extensions:
      greet: [[monitor]]

When a flow or agent trace already owns the call, the extension suppresses its own span and just runs the tool.

Metrics and traces are served by a monitoring backend plugin, registered under monitoring_module. Without one, the runtime uses a no-op default and there is nothing to query. To build a backend, see Author a monitoring backend.

Get started

Concepts

Guides

Marketplace

Contributing

Observe

Query metrics

List and read traces

Trace a standalone tool call

See also

​Query metrics

​List and read traces

​Trace a standalone tool call

​See also

Query metrics

List and read traces

Trace a standalone tool call

See also