{"data":{"slug":"future-agi-traceai","name":"traceAI","tagline":"Open-source observability for AI applications","github_url":"https://github.com/future-agi/traceAI","owner":"future-agi","repo":"traceAI","owner_avatar_url":"https://avatars.githubusercontent.com/u/147392366?v=4","primary_language":"Python","stars":221,"forks":44,"topics":["ai","ai-agents","langchain","large-language-models","observability","openai","opentelemetry","tracing"],"archived":false,"github_pushed_at":"2026-08-27T11:55:50+00:00","maintenance_label":"Active","stars_delta_30d":9,"url":"https://www.graphcanon.com/tools/future-agi-traceai","markdown_url":"https://www.graphcanon.com/tools/future-agi-traceai.md","api_url":"https://www.graphcanon.com/api/graphcanon/tools/future-agi-traceai","graph_url":"https://www.graphcanon.com/api/graphcanon/graph?tool=future-agi-traceai","description":"Open Source AI Tracing Framework built on Opentelemetry for AI Applications and Frameworks","homepage_url":"https://app.futureagi.com","license":"Apache-2.0","open_issues":15,"watchers":2,"ai_summary":"traceAI is an open-source AI tracing framework built on OpenTelemetry for AI applications and frameworks, enabling structured tracing of LLM calls, prompts, token counts, retrieval steps, and agent decisions.","readme_excerpt":"<div align=\"center\">\n\n\n\n# traceAI\n\n**Open-source observability for AI applications - trace every LLM call, prompt, token, retrieval step, and agent decision.**\n\nBuilt on [OpenTelemetry](https://opentelemetry.io/), traceAI sends structured traces to any OTel-compatible backend (Datadog, Grafana, Jaeger, Future AGI, and more). No new vendor. No new dashboard.\n\n\n\n\n\n\n\n\n\n\n\n\n[Documentation](https://docs.futureagi.com/integrations/overview) • [Examples](https://docs.futureagi.com/cookbook/cookbook8/How-To-Implement-Observability) • [Slack](https://join.slack.com/t/future-agi/shared_invite/zt-3gqwrdf2p-4oj1LVPqkQIoiS_OSrFL2A) • [PyPI](https://pypi.org/project/fi-instrumentation-otel/) • [npm](https://www.npmjs.com/package/@traceai/fi-core) • [NuGet](https://www.nuget.org/packages/fi-instrumentation-otel)\n\n</div>\n\n---\n\n<div align=\"center\">\n  <img src=\"trace-repo.gif\" alt=\"traceAI Demo\" width=\"70%\" />\n</div>\n\n---\n\n## What is traceAI?\n\nYour agent calls an LLM, retrieves context, invokes a tool, and returns an answer. When that answer is wrong, you need to know exactly where it broke - which retrieval missed, which tool returned stale data, which prompt drifted.\n\n**traceAI** captures every LLM call, prompt, token count, retrieval step, and agent decision as structured OpenTelemetry traces. Your traces live natively in Datadog, Grafana, Future AGI, Jaeger, or any OTel-compatible backend. No new vendor. No new dashboard.\n\n- **Drop-in instrumentation** for 50+ AI frameworks across 4 languages\n- **OpenTelemetry-native** - works with any OTel-compatible backend\n- **Semantic conventions** for LLM calls, agents, tools, retrieval, and vector databases\n- **Python, TypeScript, Java, and C#** with consistent APIs\n\n---\n\n## Table of Contents\n\n- [Key Features](#key-features)\n- [Quickstart](#quickstart)\n  - [Python](#python-quickstart)\n  - [TypeScript](#typescript-quickstart)\n  - [Java](#java-quickstart)\n  - [C#](#c-quickstart)\n- [Supported Frameworks](#supported-frameworks)\n  - [Python](#python)\n  - [TypeScript](#typescript)\n  - [Java](#java)\n  - [C#](#c)\n- [Compatibility Matrix](#compatibility-matrix)\n- [Architecture](#architecture)\n- [Roadmap](#roadmap)\n- [Contributing](#contributing)\n- [Contributors](#contributors)\n- [Resources](#resources)\n- [Connect With Us](#connect-with-us)\n\n## Key Features\n\n| Feature | Description |\n|---------|-------------|\n| **Standardized Tracing** | Maps AI workflows to consistent OpenTelemetry spans and attributes |\n| **Drop-in Setup** | Add 3 lines to your existing code - no refactoring needed |\n| **Multi-Framework** | 50+ integrations across Python, TypeScript, Java, and C# |\n| **Vendor Agnostic** | Works with any OpenTelemetry-compatible backend |\n| **Rich Context** | Captures prompts, completions, tokens, model params, tool calls, and more |\n| **Production-grade** | Async support, streaming, error handling, and low-overhead tracing |\n\n---\n\n## Quickstart\n\n### Python Quickstart\n\n**1. Install**\n```bash\npip install traceai-openai\n```\n\n**2. Instrument your application**\n```python\nimport os\nfrom fi_instrumentation import register\nfrom fi_instrumentation.fi_types import ProjectType\nfrom traceai_openai import OpenAIInstrumentor\nimport openai\n\n# Set up environment variables\nos.environ[\"FI_API_KEY\"] = \"<your-api-key>\"\nos.environ[\"FI_SECRET_KEY\"] = \"<your-secret-key>\"\nos.environ[\"OPENAI_API_KEY\"] = \"<your-openai-key>\"\n\n# Register tracer provider\ntrace_provider = register(\n    project_type=ProjectType.OBSERVE,\n    project_name=\"my_ai_app\"\n)\n\n# Instrument OpenAI\nOpenAIInstrumentor().instrument(tracer_provider=trace_provider)\n\n# Use OpenAI as normal - traces are captured automatically\nresponse = openai.chat.completions.create(\n    model=\"gpt-4.1\",\n    messages=[{\"role\": \"user\", \"content\": \"Hello!\"}]\n)\n```\n\n> **Tip:** Swap `traceai-openai` for any supported framework (e.g., `traceai-langchain`, `traceai-anthropic`)\n\n---\n\n### TypeScript Quickstart\n\n**1. Install**\n```bash\nnpm install @traceai/openai @traceai/fi-core\n```\n\n**2. Instrumen","github_created_at":"2025-04-15T04:57:02+00:00","created_at":"2026-07-11T10:48:14.554044+00:00","updated_at":"2026-09-20T04:30:27.752597+00:00","categories":[{"slug":"evaluation-observability","name":"Evaluation & Observability","url":"https://www.graphcanon.com/categories/evaluation-observability","markdown_url":"https://www.graphcanon.com/categories/evaluation-observability.md","api_url":"https://www.graphcanon.com/api/graphcanon/categories/evaluation-observability"}],"tags":[{"slug":"ai","name":"ai"},{"slug":"ai-agents","name":"ai-agents"},{"slug":"langchain","name":"langchain"},{"slug":"large-language-models","name":"large-language-models"},{"slug":"observability","name":"observability"},{"slug":"openai","name":"openai"},{"slug":"opentelemetry","name":"opentelemetry"},{"slug":"tracing","name":"tracing"}],"trust":{"provenance":{"is_fork":false,"github_id":966544022,"owner_type":"Organization","methodology":"github_public_v1","parent_repo":null,"near_duplicate_slugs":[]},"computed_at":"2026-09-18T08:23:37.338Z","maintenance":{"label":"Active","score":82,"methodology":"github_public_v1","releases_90d":0,"days_since_push":21,"last_release_at":"2026-03-11T14:40:15Z","stars_delta_30d":9,"open_issues_delta_30d":4},"security_summary":{"status":"no_lockfile","scanner":null,"low_count":0,"high_count":0,"last_scan_at":"2026-09-18T08:23:39.540Z","medium_count":0,"scan_profile":"none","critical_count":0}},"capability_facts":{"scan":{"source":"repo_scan","observed_at":"2026-09-18T08:23:36.242Z"},"languages":{"value":["python"],"source":"github.language","observed_at":"2026-09-18T08:23:36.242Z"},"license_spdx":{"value":"Apache-2.0","source":"github.license","observed_at":"2026-09-18T08:23:36.242Z"}},"decision_facts":{"hosting":null,"pricing":{"model":"freemium","summary":"traceAI is free and open-source, but users may incur costs based on the backend services they choose to integrate with."},"requirements":{"notes":["Python, TypeScript, Java, and C# are supported.","Requires OpenTelemetry-compatible backend for trace storage and visualization."],"min_ram_gb":2,"requires_docker":false},"constraints":{"min_ram_gb":2,"pricing_model":"freemium","requires_docker":false},"when_to_use":["When you need to trace every LLM call, prompt, token count, retrieval step, and agent decision in your AI application.","If you are working with multiple AI frameworks across Python, TypeScript, Java, and C#, and require consistent APIs for instrumentation.","When you want to integrate with any OpenTelemetry-compatible backend without the need for a new vendor or dashboard.","If your project requires production-grade features such as async support, streaming, error handling, and low-overhead tracing."],"when_not_to_use":["If your project does not require detailed tracing of LLM calls, prompts, token counts, retrieval steps, and agent decisions.","When you are working exclusively with a single AI framework that does not require cross-language consistency.","If you are looking for a solution that is not open-source or does not support OpenTelemetry.","When you prefer a vendor-specific solution over a vendor-agnostic approach."],"source":"enrich:decision_facts","observed_at":"2026-09-18T11:34:42.091Z"},"constraint_facets":{"min_ram_gb":2,"pricing_model":"freemium","requires_docker":false},"decision_summary":[{"label":"Pricing","value":"freemium - traceAI is free and open-source, but users may incur costs based on the backend services they choose to integrate with."},{"label":"Requirements","value":"Min 2 GB RAM; Python, TypeScript, Java, and C# are supported.; Requires OpenTelemetry-compatible backend for trace storage and visualization."},{"label":"Adopt for","value":"traceAI is an open-source observability framework for AI applications, built on OpenTelemetry, that provides structured tracing for LLM calls, prompts, token counts, retrieval steps, and agent decisions."},{"label":"License detail","value":"traceAI is available under the Apache-2.0 license, allowing for free use, modification, and distribution."}]}}