annapurna-meter

annapurna-meter (Node)

The optional metering hook for Annapurna — a thin, fail-safe wrapper that reports per-call LLM usage so spend can be attributed per feature. No dependencies (Node ≥ 18). Cost is computed server-side from Annapurna's pricing tables — the SDK never sees prices, and it never sends prompt or response content, only token counts and a featureId.

It is fail-safe: reporting is fire-and-forget and can never throw into your request path. With no ingest URL/token configured, every call is a no-op.

Install

npm install annapurna-meter

Use

Recommended — wrap the client once (no per-call code):

import { wrap } from "annapurna-meter";

const client = wrap(openai, { featureId: "feature-threat-triage" }); // reads ENV

const resp = await client.chat.completions.create({ model: "gpt-4o", ... }); // metered

Provider is auto-detected; each call is recorded with its latency. Pass an optional metadata (e.g. { environment, customer_id }) for extra attribution. Streaming responses use the explicit form below.

Explicit — one line per call:

import { Meter } from "annapurna-meter";

const meter = new Meter("feature-threat-triage");
const resp = await openai.chat.completions.create({ model: "gpt-4o", ... });
meter.recordOpenAI(resp);        // <- the whole hook

Helpers: recordAnthropic, recordOpenAI, plus the generic record({ provider, model, tokensIn, tokensOut, featureId }).

Optimize mode (opt-in)

new Meter(featureId, { optimize: true }) additionally emits privacy-safe signals — salted hashes and counts, never prompt text — so Annapurna can surface measured optimization opportunities (duplicate calls, uncached repeated prefixes). Off by default; work is off the call path, memory-bounded, and fail-safe. The SDK fetches a per-tenant salt once (GET /api/hook/salt, same ingest token).

Config

License

Apache-2.0. (The Annapurna server is AGPL-3.0; this client SDK is permissive so you can embed it in a proprietary app.)