Orq MCP is live: Use natural language to interrogate traces, spot regressions, and experiment your way to optimal AI configurations. Available in Claude Desktop, Claude Code, Cursor, and more. Start now →
Attach one or more annotations to a specific span. A standard annotation references a human review by key and supplies a value. A correction references an existing evaluator output by parent_annotation_id and supplies the corrected value, validated against that evaluator’s output schema.
from orq_ai_sdk import Orqimport oswith Orq( api_key=os.getenv("ORQ_API_KEY", ""),) as orq: orq.annotations.create(trace_id="<id>", span_id="<id>", annotations=[]) # Use the SDK ...
import { Orq } from "@orq-ai/node";const orq = new Orq({ apiKey: process.env["ORQ_API_KEY"] ?? "",});async function run() { await orq.annotations.create({ traceId: "<id>", spanId: "<id>", });}run();
Remove one or more annotations from a specific span by their evaluator keys, or remove corrections by the eval ids of their parent annotations.
from orq_ai_sdk import Orqimport oswith Orq( api_key=os.getenv("ORQ_API_KEY", ""),) as orq: orq.annotations.delete(trace_id="<id>", span_id="<id>") # Use the SDK ...
import { Orq } from "@orq-ai/node";const orq = new Orq({ apiKey: process.env["ORQ_API_KEY"] ?? "",});async function run() { await orq.annotations.delete({ traceId: "<id>", spanId: "<id>", });}run();