v0.3.0

What's new

OTLP export for production tracing

agent-trace sessions can now be exported as OpenTelemetry spans to any OTLP-compatible backend. Zero new dependencies.

# Honeycomb
agent-strace export <session-id> --format otlp \
  --endpoint [https://api.honeycomb.io](https://api.honeycomb.io) \
  --header "x-honeycomb-team: $HONEYCOMB_API_KEY"

# Datadog
agent-strace export <session-id> --format otlp \
  --endpoint [http://localhost:4318](http://localhost:4318)

# New Relic
agent-strace export <session-id> --format otlp \
  --endpoint [https://otlp.nr-data.net](https://otlp.nr-data.net) \
  --header "api-key: $NEW_RELIC_LICENSE_KEY"

# Splunk
agent-strace export <session-id> --format otlp \
  --endpoint https://ingest.<realm>.signalfx.com \
  --header "X-SF-Token: $SPLUNK_ACCESS_TOKEN"

# Dump OTLP JSON without sending
agent-strace export <session-id> --format otlp > trace.json

How it maps

agent-trace	OpenTelemetry
session	trace
tool_call + tool_result	span (with duration)
error	span with error status + exception event
user_prompt	event on root span
assistant_response	event on root span
session_id	trace ID
event_id	span ID
parent_id	parent span ID

In your dashboard, a session looks like a trace waterfall: root span for the session, child spans for each tool call with timing and inputs, error spans highlighted in red.

See examples/production_tracing.md for setup guides for each backend.

Install / upgrade

uv tool upgrade agent-strace
pip install agent-strace==0.3.0

130 tests. Zero dependencies. Python 3.10+ stdlib only.