ratel-ai-telemetry 0.1.0rc4

Ratel telemetry conventions — the ratel.* overlay on OpenTelemetry gen_ai spans. init() helper and constants for the Ratel context engineering platform.

ratel-ai-telemetry (Python)

The ratel.* telemetry vocabulary for Python: the constants that codify the Tier 2 overlay of ../CONVENTIONS.md (attribute keys, span/event names, the Origin/SearchTarget/AuthOutcome value enums, the pinned semconv version). Importing the constants pulls no OpenTelemetry SDK — the vocabulary stays weight-free for the SDK (emit side), the server (read side), and edge/serverless emitters (ADR-0015). init() — turnkey OTLP exporter sugar over the standard OTel Python SDK — lives in the ratel_ai_telemetry.otlp submodule behind the optional [otlp] extra.

Usage

from opentelemetry import trace
from ratel_ai_telemetry import EXECUTE_TOOL, GEN_AI_OPERATION_NAME, GEN_AI_TOOL_NAME, RATEL_ORIGIN, Origin

# Emit a standard gen_ai `execute_tool` span enriched with the ratel.* overlay,
# on your own OTel provider — the constants alone, no extra needed.
span = trace.get_tracer("my-agent").start_span(
    EXECUTE_TOOL,
    attributes={
        GEN_AI_OPERATION_NAME: EXECUTE_TOOL,
        GEN_AI_TOOL_NAME: "send_email",
        RATEL_ORIGIN: Origin.AGENT.value,
    },
)
span.end()

Want turnkey OTLP export to Ratel? Install ratel-ai-telemetry[otlp] and call init():

from ratel_ai_telemetry.otlp import init  # also importable as `from ratel_ai_telemetry import init`

provider = init(api_key="sk-...")  # wires the OTLP exporter to RATEL_URL (or endpoint=/headers=)
# ... emit spans ...
provider.shutdown()  # flush the exporter on exit

init() owns the global provider, so it is the turnkey path for a greenfield app; it raises — pointing at ratel_span_processor — rather than silently no-op'ing if a provider is already registered. A complete, offline-runnable version (console exporter + a ratel.search → execute_tool trace) is in examples/telemetry-python.

Coexisting with another provider (Langfuse, the Vercel AI SDK, ...)

OpenTelemetry's model is one provider with many span-processors. When a partner already owns the provider, add ratel_span_processor to it instead of calling init() — Ratel ingests only the gen_ai.* / ratel.* signal (the default ratel_signal_filter), dropping the framework's ai.* wrapper noise:

from opentelemetry.sdk.trace import TracerProvider
from ratel_ai_telemetry.otlp import ratel_span_processor

provider = TracerProvider()
provider.add_span_processor(existing_langfuse_processor)              # keeps every span
provider.add_span_processor(ratel_span_processor(api_key="sk-..."))  # gen_ai.*/ratel.* only

Pass span_filter=lambda _s: True (or your own predicate) to override the default; ratel_span_exporter() is the bare OTLP exporter if you want to wire your own processor. Note that per-span filtering can orphan the AI SDK's ai.* wrapper from its gen_ai.* child; send everything (or tail-sample) when you need full-trace fidelity rather than just the gen_ai/ratel metrics.

Package shape

• Distribution name: ratel-ai-telemetry; import name: ratel_ai_telemetry
• Pure Python (hatchling build, no Rust extension); OTel-free constants, init() behind the [otlp] extra
• Targets Python >=3.9 (the [otlp] OTel deps are pinned below 1.42, the last line supporting 3.9)
• Released under the telemetry-py-v* tag prefix (ADR-0016)
• MIT (ADR-0017)

Build & test

From this directory (needs uv):

uv venv --python 3.11 .venv
uv pip install --python .venv -e '.[dev]'
.venv/bin/ruff check . && .venv/bin/mypy ratel_ai_telemetry && .venv/bin/pytest

Unlike the Python SDK there is no maturin develop step — the package is pure Python, installed editable ([dev] pulls the [otlp] extra so the tests exercise the real SDK). The tests cover the vocabulary (each constant asserted against the pin), init()'s endpoint/auth resolution, its already-registered guard and the content-capture gate, the ratel_signal_filter predicate and that ratel_span_processor forwards only the spans it passes, a purity guard that importing the package pulls no OTel, and the shared contract-against-the-pin conformance in ../conformance/ (spans built from these constants through the real SDK must emit the exact pinned keys).