freshcrate
Skin:/
Language:English/įŽ€äŊ“中文
Home > Frameworks > monocle

monocle

Monocle is a framework for tracing GenAI app code. This repo contains implementation of Monocle for GenAI apps written in Python.

generative-ailinux-foundationllm-agentllm-inferencellmsobservabilityopentelemetryosspython
Why this rank:Strong adoptionRecent releaseHealthy release cadence

Description

Monocle is a framework for tracing GenAI app code. This repo contains implementation of Monocle for GenAI apps written in Python.

README

Monocle for tracing GenAI app code

PyPI Version Discord Slack

Monocle helps developers and platform engineers building or managing GenAI apps monitor these in prod by making it easy to instrument their code to capture traces that are compliant with open-source cloud-native observability ecosystem.

Monocle is a community-driven OSS framework for tracing GenAI app code governed as a Linux Foundation AI & Data project.

Why Monocle

Monocle is built for:

  • app developers to trace their app code in any environment without lots of custom code decoration
  • platform engineers to instrument apps in prod through wrapping instead of asking app devs to recode
  • GenAI component providers to add observability features to their products
  • enterprises to consume traces from GenAI apps in their existing open-source observability stack

Benefits:

  • Monocle provides an implementation + package, not just a spec
    • No expertise in OpenTelemetry spec required
    • No bespoke implementation of that spec required
    • No last-mile GenAI domain specific code required to instrument your app
  • Monocle provides consistency
    • Connect traces across app code executions, model inference or data retrievals
    • No cleansing of telemetry data across GenAI component providers required
    • Works the same in personal lab dev or org cloud prod environments
    • Send traces to location that fits your scale, budget and observability stack
  • Monocle is fully open source and community driven
    • No vendor lock-in
    • Implementation is transparent
    • You can freely use or customize it to fit your needs

What Monocle provides

  • Easy to use code instrumentation
  • OpenTelemetry compatible format for spans.
  • Community-curated and extensible metamodel for consisent tracing of GenAI components.
  • Export to local and cloud storage

Use Monocle

Generate traces

Install monocle package and make simple two line change to your application to generate traces.

    pip install monocle_apptrace
  • Instrument your app code
    • Import the Monocle package 
         from monocle_apptrace.instrumentor import setup_monocle_telemetry
    • Setup instrumentation in your main() function 
         setup_monocle_telemetry(workflow_name="your-app-name")
  • (Optionally) Modify config to alter where traces are sent

See Monocle user guide for more details.

Monocle test tool

Test your application and agent behavior by using Monocle's genAI test tool. Just define your input, expected output and expected agent or tool to be called. The Monocle test tool automatically generates the traces and validate your expected behavior.

Monocle MCP server

The MCP server provided by Monocle integrates with your dev enviorment like Visual Studio and Github copilot. It provides curated prompts and tools to analyze the trace to find issues.

Roadmap

Goal of Monocle is to support tracing for apps written in any language with any LLM orchestration or agentic framework and built using models, vectors, agents or other components served up by any cloud or model inference provider.

Current version supports:

  • Language: (đŸŸĸ) Python , (đŸŸĸ) Typescript
  • Agentic frameworks: (đŸŸĸ) Langgraph, (đŸŸĸ) LlamaIndex, (đŸŸĸ) Google ADK, (đŸŸĸ) OpenAI Agent SDK, (đŸŸĸ) AWS Strands, (đŸŸĸ) CrewAI, (đŸŸĸ) Microsoft Agent Framework
  • MCP/A2A frameworks: (đŸŸĸ) FastMCP, (đŸŸĸ) MCP client, (đŸŸĸ) A2A client
  • Web/App frameworks: (đŸŸĸ) Flask, (đŸŸĸ) AIO Http, (đŸŸĸ)FastAPI, (đŸŸĸ) Azure Function, (đŸŸĸ) AWS Lambda, (đŸŸĸ) Vercel (typescript), (đŸŸĸ) Microsoft Teams AI SDK, (đŸŸĸ) Web/REST client, (🔜) Google Function,
  • LLM-frameworks: (đŸŸĸ) Langchain, (đŸŸĸ) Llamaindex, (đŸŸĸ) Haystack
  • Agent Runtime: (đŸŸĸ) AWS Bedrock Agentcore
  • LLM inference providers: (đŸŸĸ) OpenAI, (đŸŸĸ) Azure OpenAI, (đŸŸĸ) Azure AI, (đŸŸĸ) Nvidia Triton, (đŸŸĸ) AWS Bedrock, (đŸŸĸ) AWS Sagemaker, (đŸŸĸ) Google Vertex, (đŸŸĸ) Google Gemini, (đŸŸĸ) Hugging Face, (đŸŸĸ) Deepseek, (đŸŸĸ) Anthropic, (đŸŸĸ) Mistral, (đŸŸĸ) LiteLLM ,(🔜) Azure ML
  • Vector stores: (đŸŸĸ) FAISS, (🔜) OpenSearch, (🔜) Milvus
  • Exporter: (đŸŸĸ) stdout, (đŸŸĸ) file, (đŸŸĸ) Memory, (đŸŸĸ) Azure Blob Storage, (đŸŸĸ) AWS S3, (đŸŸĸ) Okahu cloud, (đŸŸĸ) OTEL compatible collectors, (đŸŸĸ) Google Cloud Storage

Get involved

Provide feedback

  • Submit issues and enhancements requests via Github issues

Contribute

  • Monocle is community based open source project. We welcome your contributions. Please refer to the CONTRIBUTING and CODE_OF_CONDUCT for guidelines. The contributor's guide provides technical details of the project.

Release History

VersionChangesUrgencyDate
v0.8.3This is a patch release on the previous 0.8.2 release, fixing the issue(s) below. - feat: Add Okahu Cloud Browser Sign-In for Monocle Agentic CLI Setup ([#592](https://github.com/monocle2ai/monocle/pull/592)) - fix: Handle MS Agent SDK invocation span and missing inference events ([#593](https://github.com/monocle2ai/monocle/pull/593)) - fix: Update tool type for inference span to generic tool.function in OpenAI ([#596](https://github.com/monocle2ai/monocle/pull/596)) - fix: Incorrect span subHigh6/3/2026
v0.8.2This is a patch release on the previous 0.8.1 release, fixing the issue(s) below. - Fix test framework issue [584](https://github.com/monocle2ai/monocle/pull/584) High5/27/2026
v0.8.1This is a patch release on the previous 0.8.0 release, fixing the issue(s) below. - Codex instrumentation and UV packaging ([#538](https://github.com/monocle2ai/monocle/pull/538)) - Fetch traces by scope ID ([#542](https://github.com/monocle2ai/monocle/pull/542)) - Handle tool output being dict/json ([#544](https://github.com/monocle2ai/monocle/pull/544)) - Span subtypes in ADK ([#546](https://github.com/monocle2ai/monocle/pull/546)) - Set inference.decision.span.id for ADK ([#548](https://gitHigh5/19/2026
v0.8.0- Claude code instrumentation - Custom span enhancement - capture arguement and results High4/30/2026
v0.7.9This is a patch release on the previous 0.7.8 release, fixing the issue(s) below. - Monocle test framework to support Okahu eval testing for facts beyond just traces - Okahu exporter to load API key from .env.monocle if provided - Minor bugfixes High4/27/2026
v0.7.8This is a patch release on the previous 0.7.7 release, fixing the issue(s) below. - Updated langrapgh dependencies to latest versions ([#517](https://github.com/monocle2ai/monocle/pull/517)) - Add execution ID for parellel agent runs ([#488](https://github.com/monocle2ai/monocle/pull/488)) - Implement paygentic exporter ([#501](https://github.com/monocle2ai/monocle/pull/501)) - Hugging face streaming support ([#506](https://github.com/monocle2ai/monocle/pull/506)) - Support for filtering speciHigh4/15/2026
v0.7.7This is a patch release on the previous 0.7.6 release, fixing the issue(s) below. - Implement MS Agent streaming support ([#487](https://github.com/monocle2ai/monocle/pull/487)) - Implement Azure and OpenAI specific stream processors ([#474](https://github.com/monocle2ai/monocle/pull/474)) - Support under_duration assertion in monocle test frameworks ([#484](https://github.com/monocle2ai/monocle/pull/484)) - MS Agent with OpenAI inference integration ([#483](https://github.com/monocle2ai/monocMedium3/21/2026
v0.7.6This is a patch release on the previous 0.7.5 release, fixing the issue(s) below. - Updated runners to include session id details ([#461](https://github.com/monocle2ai/monocle/pull/461)) - Added strands runner and tests in test_tools with session_id ([#459](https://github.com/monocle2ai/monocle/pull/459)) - Feature: Pytest Evaluations ([#467](https://github.com/monocle2ai/monocle/pull/467)) - Fix for session id in msagent ([#460](https://github.com/monocle2ai/monocle/pull/460)) Low2/27/2026
v0.7.5This is a patch release on the previous 0.7.4 release, fixing the issue(s) below. - Add dependency for agent-framework-core ([#455](https://github.com/monocle2ai/monocle/pull/455)) - Ensure that the fluent trace asserter has full set of spans at start of every chain ([#453](https://github.com/monocle2ai/monocle/pull/453)) - Add custom workflow name for test tool ([#454](https://github.com/monocle2ai/monocle/pull/454)) - Add for flask and aiohttp skip empty heathcheck http spans ([#448](https:/Low2/19/2026
v0.7.4This is a patch release on the previous 0.7.3 release, fixing the issue(s) below. - Refactor span export logic in okahu_exporter.py ([#449](https://github.com/monocle2ai/monocle/pull/449)) - gcs exporter ([#429](https://github.com/monocle2ai/monocle/pull/429)) - Fixed session_id/chat_store_key extraction in lmx and updated test ([#440](https://github.com/monocle2ai/monocle/pull/440)) - Changed tool method ([#442](https://github.com/monocle2ai/monocle/pull/442)) - Fix merge resolution issues inLow2/11/2026
v0.7.3This is a patch release on the previous 0.7.2 release, fixing the issue(s) below. - Scope attributes MS Agent ([#428](https://github.com/monocle2ai/monocle/pull/428)) - Add input and output for aiohttp MSAgent SDK ([#425](https://github.com/monocle2ai/monocle/pull/425)) - Update boto3 version to 1.40.52 and adjust OpenSearch endpoint variable for SageMaker integration ([#431](https://github.com/monocle2ai/monocle/pull/431)) - Add attribute in retrieval span for KnowledgeBase ([#427](https://giLow1/29/2026
v0.7.2This is a patch release on the previous 0.7.1 release, fixing the issue(s) below. - AWS Agentcore support (#412) - Add Boto Coverage to Support API (#411) - Microsoft Agent Framework support for monocle (#401) - Improve S3SpanExporter export method to handle existing event loops (#408) - Update extract_tool_response for MCP CallToolResult type (#405) - Added links to Discord and Slack (#403) - Fix the response attribute in agentic tool for langgraph (#402) - Cleanup the class member var correcLow1/20/2026
v0.7.1This is a patch release on the previous 0.7.0 release, fixing the issue(s) below. - Update extract_tool_response for MCP CallToolResult type ([#405](https://github.com/monocle2ai/monocle/pull/405)) - Added links to Discord and Slack ([#403](https://github.com/monocle2ai/monocle/pull/403)) - Fix the response attribute in agentic tool for langgraph ([#402](https://github.com/monocle2ai/monocle/pull/402)) - Refactor and update version in pyproject.toml; fix instrumentor setup ([#399](https://githLow1/12/2026
v0.7.0 - Added fix for Strand Boto Input ([#365](https://github.com/monocle2ai/monocle/pull/365)) - Added fix for Failed to Entity index ([#363](https://github.com/monocle2ai/monocle/pull/363)) - Handle test status and skipping passed test trace exports ([#362](https://github.com/monocle2ai/monocle/pull/362)) - Fix typo in strand session id attribute ([#359](https://github.com/monocle2ai/monocle/pull/359)) - CrewAI metamodel and tests ([#343](https://github.com/monocle2ai/monocle/pull/343)) - Basic flLow12/8/2025
v0.6.6This is a patch release on the previous 0.6.5 release, fixing the issue(s) below. - Handle langgraph request input ([#321](https://github.com/monocle2ai/monocle/pull/321)) - fastapi methods update to fix coroutine issue during server start ([#322](https://github.com/monocle2ai/monocle/pull/322)) Low11/5/2025
v0.6.5This is a patch release on the previous 0.6.4 release, fixing the issue(s) below. - Added scope agentic invocation ([#282](https://github.com/monocle2ai/monocle/pull/282)) - Standardize the inference sub type similar to other span.subtypes ([#303](https://github.com/monocle2ai/monocle/pull/303)) - Fix mcp package name in Readme ([#300](https://github.com/monocle2ai/monocle/pull/300)) - Fix llamaindex agent and Llamaindex multi_agent ([#309](https://github.com/monocle2ai/monocle/pull/309)) - FLow11/3/2025
v0.6.0 - Monocle testing framework ([#288](https://github.com/monocle2ai/monocle/pull/288)) - Fix package version conflict ([#293](https://github.com/monocle2ai/monocle/pull/293)) - Remove hugging face direct dependency on Monocle. Also update the tests - Hugging face inference test ([#290](https://github.com/monocle2ai/monocle/pull/290)) - Attributes fixes ([#289](https://github.com/monocle2ai/monocle/pull/289)) - mistral embed instrumentation ([#279](https://github.com/monocle2ai/monocle/pull/279)) Low10/1/2025
v0.5.3This is a patch release on the previous 0.5.2 release, fixing the issue(s) below. - Make ./.monocle as default folder to store traces instead of current directory ([#272](https://github.com/monocle2ai/monocle/pull/272)) - HTTP span formatting ([#258](https://github.com/monocle2ai/monocle/pull/258)) - Adding mistral metamodel - instrumentation of inference api ([#273](https://github.com/monocle2ai/monocle/pull/273)) - Add Http Url for Azfunc, Lambda, Flask and FastApi Metamodel ([#254](https://Low9/9/2025
v0.5.2This is a patch release on the previous 0.5.1 release, fixing the issue(s) below. - Add monocle MCP server ([#264](https://github.com/monocle2ai/monocle/pull/264)) Low8/26/2025
v0.5.1This is a patch release on the previous 0.5.0 release, fixing the issue(s) below. - Support injecting synthetic spans and use that for adding delegation span in Google ADK ([#257](https://github.com/monocle2ai/monocle/pull/257)) - Added monocle subtypes ([#256](https://github.com/monocle2ai/monocle/pull/256)) - OpenAI agents request span ([#255](https://github.com/monocle2ai/monocle/pull/255)) - Set scopes for agent request and invocation plus minor fixes ([#253](https://github.com/monocle2ai/Low8/25/2025
v0.5.0 - Fix missing inference span in google ADK ([#249](https://github.com/monocle2ai/monocle/pull/249)) - Added openai agents instrumentation ([#248](https://github.com/monocle2ai/monocle/pull/248)) - Google Agent development kit meta mode ([#247](https://github.com/monocle2ai/monocle/pull/247)) - Add Support for LiteLLM OpenAI and AzureOpenAI ([#246](https://github.com/monocle2ai/monocle/pull/246)) - Added inference subtype for langgrap, openai and anthropic ([#245](https://github.com/monocle2ai/mLow8/13/2025
v0.5.0b1 - Add support for teams finish type, move finish types to metadata section ([#240](https://github.com/monocle2ai/monocle/pull/240)) - Fix integration tests ([#239](https://github.com/monocle2ai/monocle/pull/239)) - Add function_name attribute for Azure function ([#238](https://github.com/monocle2ai/monocle/pull/238)) - Added MCP and A2A ([#237](https://github.com/monocle2ai/monocle/pull/237)) - Add Metamodel for Lambda Func ([#236](https://github.com/monocle2ai/monocle/pull/236)) - Allow llm SDLow8/4/2025
v0.4.2This is a patch release on the previous 0.4.1 release, fixing the issue(s) below. - Add gemini instrumentation ([#220](https://github.com/monocle2ai/monocle/pull/220)) Low6/26/2025
v0.4.1This is a patch release on the previous 0.4.0 release, fixing the issue(s) below. - Add exception status code for Boto3 ([#211](https://github.com/monocle2ai/monocle/pull/211)) - Add exception status code for anthropic and openai ([#210](https://github.com/monocle2ai/monocle/pull/210)) - Add prompt template info in teamsAI ([#209](https://github.com/monocle2ai/monocle/pull/209)) - TeamsAI : added system prompt ([#208](https://github.com/monocle2ai/monocle/pull/208)) - Add prompt template info Low6/18/2025
v0.4.0 - Update teams scopes ([#200](https://github.com/monocle2ai/monocle/pull/200)) - Record input and errors for inference.modelapi in case of error ([#193](https://github.com/monocle2ai/monocle/pull/193)) - Removed special handling for streaming in wrapper ([#192](https://github.com/monocle2ai/monocle/pull/192)) - Add Span error handling ([#186](https://github.com/monocle2ai/monocle/pull/186)) - Add teams ai enhancements ([#184](https://github.com/monocle2ai/monocle/pull/184)) - Added conversaLow6/3/2025
v0.4.0b3 - Record input and errors for inference.modelapi in case of error ([#193](https://github.com/monocle2ai/monocle/pull/193)) - Removed special handling for streaming in wrapper ([#192](https://github.com/monocle2ai/monocle/pull/192)) Low5/28/2025
v0.4.0b2 - Add Span error handling ([#186](https://github.com/monocle2ai/monocle/pull/186)) - Add teams ai enhancements ([#184](https://github.com/monocle2ai/monocle/pull/184)) Low5/21/2025
v0.4.0b1 - Added conversation id in scope for teams ai bot ([#180](https://github.com/monocle2ai/monocle/pull/180)) - Update inference entity type of TeamsAI SDK ([#178](https://github.com/monocle2ai/monocle/pull/178)) - Added stream and async for openai ([#177](https://github.com/monocle2ai/monocle/pull/177)) - Update inference span of TeamsAI ([#176](https://github.com/monocle2ai/monocle/pull/176)) - Remove Preset span name and Bugfix for Event ([#175](https://github.com/monocle2ai/monocle/pull/175)) Low5/14/2025
v0.3.1This is a patch release on the previous 0.3.0 release, fixing the issue(s) below. - Add MetaModel for Anthropic SDK ([#159](https://github.com/monocle2ai/monocle/pull/159)) - Add openAI response for openAI and AzureOpenAI ([#158](https://github.com/monocle2ai/monocle/pull/158)) - Update retrieval span for Boto Client ([#157](https://github.com/monocle2ai/monocle/pull/157)) - Resolve token threshold error ([#156](https://github.com/monocle2ai/monocle/pull/156)) - Update Inference Span ([#155](hLow4/18/2025
v0.3.1b1This is a patch release on the previous 0.3.0 release, fixing the issue(s) below. - Add MetaModel for Anthropic SDK ([#159](https://github.com/monocle2ai/monocle/pull/159)) - Add openAI response for openAI and AzureOpenAI ([#158](https://github.com/monocle2ai/monocle/pull/158)) - Update retrieval span for Boto Client ([#157](https://github.com/monocle2ai/monocle/pull/157)) - Resolve token threshold error ([#156](https://github.com/monocle2ai/monocle/pull/156)) - Update Inference Span ([#155](hLow4/15/2025
v0.3.0 - Fixed issue with passing context in async case ([#150](https://github.com/monocle2ai/monocle/pull/150)) - Added lambda processor ([#148](https://github.com/monocle2ai/monocle/pull/148)) - Setup package level init scripts to make the monocle import simpler ([#147](https://github.com/monocle2ai/monocle/pull/147)) - Boto attributes and test cleanup ([#146](https://github.com/monocle2ai/monocle/pull/146)) - Openai workflow ([#142](https://github.com/monocle2ai/monocle/pull/142)) - Add input/outpuLow3/18/2025
v0.3.0b7 - Add dev dependency for Mistral AI integration ([#81](https://github.com/monocle2ai/monocle/pull/81)) - Add VectorStore deployment URL capture support ([#80](https://github.com/monocle2ai/monocle/pull/80)) - Clean up cloud exporter implementation ([#79](https://github.com/monocle2ai/monocle/pull/79)) - Capture inference span input/output events attributes ([#77](https://github.com/monocle2ai/monocle/pull/77)) - Add release automation workflows ([#76](https://github.com/monocle2ai/monocle/pulLow3/17/2025
v0.3.0b6 - Add dev dependency for Mistral AI integration ([#81](https://github.com/monocle2ai/monocle/pull/81)) - Add VectorStore deployment URL capture support ([#80](https://github.com/monocle2ai/monocle/pull/80)) - Clean up cloud exporter implementation ([#79](https://github.com/monocle2ai/monocle/pull/79)) - Capture inference span input/output events attributes ([#77](https://github.com/monocle2ai/monocle/pull/77)) - Add release automation workflows ([#76](https://github.com/monocle2ai/monocle/pulLow3/10/2025
v0.3.0b5 - Add dev dependency for Mistral AI integration ([#81](https://github.com/monocle2ai/monocle/pull/81)) - Add VectorStore deployment URL capture support ([#80](https://github.com/monocle2ai/monocle/pull/80)) - Clean up cloud exporter implementation ([#79](https://github.com/monocle2ai/monocle/pull/79)) - Capture inference span input/output events attributes ([#77](https://github.com/monocle2ai/monocle/pull/77)) - Add release automation workflows ([#76](https://github.com/monocle2ai/monocle/pulLow3/4/2025
v0.3.0b4 - Add dev dependency for Mistral AI integration ([#81](https://github.com/monocle2ai/monocle/pull/81)) - Add VectorStore deployment URL capture support ([#80](https://github.com/monocle2ai/monocle/pull/80)) - Clean up cloud exporter implementation ([#79](https://github.com/monocle2ai/monocle/pull/79)) - Capture inference span input/output events attributes ([#77](https://github.com/monocle2ai/monocle/pull/77)) - Add release automation workflows ([#76](https://github.com/monocle2ai/monocle/pulLow2/11/2025
v0.3.0b3 - Add dev dependency for Mistral AI integration ([#81](https://github.com/monocle2ai/monocle/pull/81)) - Add VectorStore deployment URL capture support ([#80](https://github.com/monocle2ai/monocle/pull/80)) - Clean up cloud exporter implementation ([#79](https://github.com/monocle2ai/monocle/pull/79)) - Capture inference span input/output events attributes ([#77](https://github.com/monocle2ai/monocle/pull/77)) - Add release automation workflows ([#76](https://github.com/monocle2ai/monocle/pulLow2/6/2025
v0.3.0b2 - Add dev dependency for Mistral AI integration ([#81](https://github.com/monocle2ai/monocle/pull/81)) - Add VectorStore deployment URL capture support ([#80](https://github.com/monocle2ai/monocle/pull/80)) - Clean up cloud exporter implementation ([#79](https://github.com/monocle2ai/monocle/pull/79)) - Capture inference span input/output events attributes ([#77](https://github.com/monocle2ai/monocle/pull/77)) - Add release automation workflows ([#76](https://github.com/monocle2ai/monocle/pulLow12/24/2024
v0.3.0b1 - Add dev dependency for Mistral AI integration ([#81](https://github.com/monocle2ai/monocle/pull/81)) - Add VectorStore deployment URL capture support ([#80](https://github.com/monocle2ai/monocle/pull/80)) - Clean up cloud exporter implementation ([#79](https://github.com/monocle2ai/monocle/pull/79)) - Capture inference span input/output events attributes ([#77](https://github.com/monocle2ai/monocle/pull/77)) - Add release automation workflows ([#76](https://github.com/monocle2ai/monocle/pulLow12/10/2024
0.2.0# Monocle 0.2.0 release ## Release artifact ### Pypi https://pypi.org/project/monocle-apptrace/0.2.0/ ## Highlights - Metamodel implementation: Monocle traces are emitted in the new metamodel format! - Store the traces directly in AWS S3 or Azure blob storage - Minor improvements and bugfixes **Full Changelog**: https://github.com/monocle2ai/monocle/compare/0.1.1...0.2.0Low11/7/2024
0.1.1# Monocle Release 0.1.1 ## Release artifact ### Pypi https://pypi.org/project/monocle-apptrace/0.1.1/ # Highlights - Metamodel specs and documentation - Improvements in vector traces - Minor improvements and bugfixes ## What's Changed * Update key for session context field in attributes by @kshitiz-okahu in https://github.com/monocle2ai/monocle/pull/34 * Add package import step to readme by @prasad-okahu in https://github.com/monocle2ai/monocle/pull/37 * improvement in vectorstoLow9/30/2024
0.1.0Release 0.1.0Low8/28/2024
0.0.1Initial Monocle release on Pypi pip install monocle_apptraceLow7/23/2024

Dependencies & License Audit

Loading dependencies...

Similar Packages

0xClawđŸĻ€ The first autonomous hackathon agent stop assisting and start competing (🏆 Hackathon Champion Project).master@2026-06-01
ToolAgentsToolAgents is a lightweight and flexible framework for creating function-calling agents with various language models and APIs.0.3.0
aitoolmanControllable and Transparent LLM Application Frameworkmain@2026-06-06
ComfyUI-LoaderUtils🔄 Optimize model loading in ComfyUI with flexible node connections and controlled sequences for better performance and memory management.main@2026-06-06
agentmarkMarkdown for the AI era@agentmark-ai/ui-components@0.6.4

More in Frameworks

schemathesisProperty-based testing framework for Open API and GraphQL based apps
spec_driven_developSpec-Driven Develop is a platform-agnostic AI agent skill that automates the pre-development workflow for large-scale complex tasks. It is not a framework, not a runtime, not a package manager — it is
DrasilGenerate all the things (focusing on research software)
langchainThe agent engineering platform