CHANGELOG.md

Changelog

All notable changes to this project will be documented in this file.

[1.1.0] - 2025-05-16

Key Features

• Full MCP (Model Context Protocol) support
• Weave tracing
• Agno integration
• ReWOO Agent
• Alert Triage Agent Example

What's Changed

• Have the examples README point to the absolute path by @AnuradhaKaruppiah in NVIDIA#4
• Set initial version will be 1.0.0 by @dagardner-nv in NVIDIA#7
• Update examples/simple_rag/README.md to verify the installation of lxml by @yczhang-nv in NVIDIA#9
• Use a separate README for pypi by @AnuradhaKaruppiah in NVIDIA#10
• Document the need to install from source to run examples by @dagardner-nv in NVIDIA#8
• Fixing broken links by @dagardner-nv in NVIDIA#14
• Cleanup readmes by @mdemoret-nv in NVIDIA#15
• Pypi readme updates by @AnuradhaKaruppiah in NVIDIA#16
• Final 1.0.0 cleanup by @AnuradhaKaruppiah in NVIDIA#18
• Add subpackage readmes redirecting to the main package by @AnuradhaKaruppiah in NVIDIA#20
• Update README.md by @gzitzlsb-nv in NVIDIA#25
• Fix #27 Documentation fix by @atalhens in NVIDIA#28
• Fix #29 - Simple_calculator example throws error - list index out of range when given subtraction by @atalhens in NVIDIA#31
• Fix: #32 Recursion Issue by @atalhens in NVIDIA#33
• "Sharing NVIDIA AgentIQ Components" docs typo fix by @avoroshilov in NVIDIA#42
• First pass at setting up issue templates by @dagardner-nv in NVIDIA#6
• Provide a cleaner progress bar when running evaluators in parallel by @AnuradhaKaruppiah in NVIDIA#38
• Setup GHA CI by @dagardner-nv in NVIDIA#46
• Switch UI submodule to https by @AnuradhaKaruppiah in NVIDIA#53
• gitlab ci pipeline cleanup by @AnuradhaKaruppiah in NVIDIA#54
• Allow str or None for retriever description by @AnuradhaKaruppiah in NVIDIA#55
• Fix case where res['categories'] = None by @balvisio in NVIDIA#22
• Misc CI improvements by @dagardner-nv in NVIDIA#56
• CI Documentation improvements by @dagardner-nv in NVIDIA#24
• Add missing platformdirs dependency by @yczhang-nv in NVIDIA#62
• Fix aiq command error when the parent directory of AIQ_CONFIG_DIR does not exist by @yczhang-nv in NVIDIA#63
• Fix broken image link in multi_frameworks documentation by @dagardner-nv in NVIDIA#61
• Updating doc string for AIQSessionManager class. by @ericevans-nv in NVIDIA#64
• Fix ragas evaluate unit tests by @AnuradhaKaruppiah in NVIDIA#68
• Normalize Gannt Chart Timestamps in Profiler Nested Stack Analysis by @dnandakumar-nv in NVIDIA#70
• Scripts for running CI locally by @dagardner-nv in NVIDIA#59
• Update types for topic and description attributes in AIQRetrieverConfig to allow None by @dagardner-nv in NVIDIA#76
• Add support for customizing output and uploading it to remote storage (S3 bucket) by @AnuradhaKaruppiah in NVIDIA#71
• Support ARM in CI by @dagardner-nv in NVIDIA#73
• Allow overriding configuration values not set in the YAML by @dagardner-nv in NVIDIA#85
• Fix bug where --workers flag was being ignored by @dagardner-nv in NVIDIA#88
• Adding Cors config for api server by @ericevans-nv in NVIDIA#89
• Update changelog for 1.1.0a1 alpha release by @AnuradhaKaruppiah in NVIDIA#90
• Updated changelog with another bug fix by @AnuradhaKaruppiah in NVIDIA#93
• Adjust how the base_sha is passed into the workflow by @dagardner-nv in NVIDIA#81
• Changes for evaluating remote workflows by @AnuradhaKaruppiah in NVIDIA#57
• Fix a bug in our pytest plugin causing test coverage to be under-reported by @dagardner-nv in NVIDIA#105
• Docker container for AgentIQ by @dagardner-nv in NVIDIA#87
• Modify JSON serialization to handle non-serializable objects by @dnandakumar-nv in NVIDIA#106
• Upload nightly builds and release builds to pypi by @dagardner-nv in NVIDIA#112
• Ensure the nightly builds have a unique alpha version number by @dagardner-nv in NVIDIA#115
• Ensure tags are fetched prior to determining the version by @dagardner-nv in NVIDIA#116
• Fix CI variable value by @dagardner-nv in NVIDIA#117
• Use setuptools_scm environment variables to set the version by @dagardner-nv in NVIDIA#118
• Only set the setuptools_scm variable when performing a nightly build by @dagardner-nv in NVIDIA#119
• Add a release PR template by @dagardner-nv in NVIDIA#123
• Add an async /evaluate endpoint to trigger evaluation jobs on a remote cluster by @AnuradhaKaruppiah in NVIDIA#109
• Update /evaluate endpoint doc by @AnuradhaKaruppiah in NVIDIA#126
• Add function tracking decorator and update IntermediateStep by @dnandakumar-nv in NVIDIA#98
• Fix typo in aiq.profiler.decorators by @dnandakumar-nv in NVIDIA#132
• Update the start command to use validate_schema by @dagardner-nv in NVIDIA#82
• Document using local/self-hosted models by @dagardner-nv in NVIDIA#101
• added Agno integration by @wenqiglantz in NVIDIA#36
• MCP Front-End Implementation by @VictorYudin in NVIDIA#133
• Make kwargs optional to the eval output customizer scripts by @AnuradhaKaruppiah in NVIDIA#139
• Add an example that shows simple_calculator running with a MCP service. by @AnuradhaKaruppiah in NVIDIA#131
• add gitdiagram to README by @yczhang-nv in NVIDIA#141
• Updating HITL reference guide to instruct users to toggle ws mode and… by @ericevans-nv in NVIDIA#142
• Add override option to the eval CLI command by @Hritik003 in NVIDIA#129
• Implement ReWOO Agent by @yczhang-nv in NVIDIA#75
• Fix type hints and docstrings for ModelTrainer by @dagardner-nv in NVIDIA#107
• Delete workflow confirmation check in CLI - #114 by @atalhens in NVIDIA#137
• Improve Agent logging by @yczhang-nv in NVIDIA#136
• Add nicer error message for agents without tools by @jkornblum-nv in NVIDIA#146
• Add colorama to core dependency by @yczhang-nv in NVIDIA#149
• Rename packages agentiq -> aiqtoolkit by @dagardner-nv in NVIDIA#152
• Rename AIQ_COMPONENT_NAME, remove unused COMPONENT_NAME by @dagardner-nv in NVIDIA#153
• Group wheels under a common aiqtoolkit directory by @dagardner-nv in NVIDIA#154
• Fix wheel upload wildcards by @dagardner-nv in NVIDIA#155
• Support Python 3.11 for AgentIQ by @yczhang-nv in NVIDIA#148
• fix pydantic version incompatibility, closes #74 by @zac-wang-nv in NVIDIA#159
• Rename AgentIQ to Agent Intelligence Toolkit by @dagardner-nv in NVIDIA#160
• Create config file symlink with aiq workflow create command by @mpenn in NVIDIA#166
• Rename generate/stream/full to generate/full and add filter_steps parameter by @AnuradhaKaruppiah in NVIDIA#164
• Add support for environment variable interpolation in config files by @mpenn in NVIDIA#157
• UI submodule rename by @AnuradhaKaruppiah in NVIDIA#168
• Consistent Trace Nesting in Parallel Function Calling by @dnandakumar-nv in NVIDIA#162
• Fix broken links in examples documentation by @dagardner-nv in NVIDIA#177
• Remove support for Python 3.13 by @yczhang-nv in NVIDIA#178
• Add transitional packages by @dagardner-nv in NVIDIA#181
• Add a tunable RAG evaluator by @liamy-nv in NVIDIA#110
• CLI Documentation fixes in remote registry configuration section by @mpenn in NVIDIA#184
• Fix uploading of transitional packages by @dagardner-nv in NVIDIA#187
• Update AIQChatRequest to support image and audio input by @yczhang-nv in NVIDIA#182
• Fix hyperlink ins the simple_calculator README by @AnuradhaKaruppiah in NVIDIA#188
• Add support for fine-grained tracing using W&B Weave by @ayulockin in NVIDIA#170
• Fix typo in CPR detected by co-pilot by @AnuradhaKaruppiah in NVIDIA#190
• Note the name change in the top-level documentation and README.md by @dagardner-nv in NVIDIA#163
• fix typo in evaluate documentation for max_concurrency by @soumilinandi in NVIDIA#191
• Fix a typo in the weave README by @AnuradhaKaruppiah in NVIDIA#195
• Update simple example eval dataset by @yczhang-nv in NVIDIA#200
• Config option to specify the intermediate step types in workflow_output.json by @AnuradhaKaruppiah in NVIDIA#198
• Update the Judge LLM settings in the examples to avoid retries by @AnuradhaKaruppiah in NVIDIA#204
• Make opentelemetry and phoenix as optional dependencies by @yczhang-nv in NVIDIA#167
• Support user-defined HTTP request metadata in workflow tools. by @ericevans-nv in NVIDIA#130
• Check if request is present before setting attributes by @AnuradhaKaruppiah in NVIDIA#209
• Add the alert triage agent example by @hsin-c in NVIDIA#193
• Updating ui submodule by @ericevans-nv in NVIDIA#211
• Fix plugin dependencies by @dagardner-nv in NVIDIA#208
• [FEA]add profiler agent to the examples folder by @zac-wang-nv in NVIDIA#120
• Regenerate uv.lock, cleaned up pyproject.toml for profiler agent example and fixed broken link in README by @yczhang-nv in NVIDIA#210
• Removed disable=unused-argument from pylint checks by @Hritik003 in NVIDIA#186
• Exception handling for discovery_metadata.py by @VictorYudin in NVIDIA#215
• Fix incorrect eval output config access by @AnuradhaKaruppiah in NVIDIA#219
• Treat a tagged commit the same as a nightly build by @dagardner-nv in NVIDIA#217
• Feature/add aiqtoolkit UI submodule by @ericevans-nv in NVIDIA#214
• Add a CLI command to list all tools available via the MCP server by @AnuradhaKaruppiah in NVIDIA#221
• For remote evaluation, workflow config is not needed by @AnuradhaKaruppiah in NVIDIA#225
• Move configurable parameters from env vars to config file by @hsin-c in NVIDIA#222
• Fix vulnerabilities in the alert triage agent example by @hsin-c in NVIDIA#227
• Add e2e test for the alert triage agent by @hsin-c in NVIDIA#226
• Fix remaining nSpect vulnerabilities for 1.1.0 by @yczhang-nv in NVIDIA#229
• Remove redundant span stack handling and error logging by @dnandakumar-nv in NVIDIA#231
• Feature/add aiqtoolkit UI submodule by @ericevans-nv in NVIDIA#234
• Fix Dockerfile build failure for v1.1.0-rc3 by @yczhang-nv in NVIDIA#240
• Bugfix for alert triage agent to run in python 3.11 by @hsin-c in NVIDIA#244
• Misc example readme fixes by @dagardner-nv in NVIDIA#242
• Fix multiple documentation and logging bugs for v1.1.0-rc3 by @yczhang-nv in NVIDIA#245
• Consolidate MCP client and server docs, examples by @AnuradhaKaruppiah in NVIDIA#246
• Update version of llama-index to 0.12.21 by @dagardner-nv in NVIDIA#257
• Fix environment variable interpolation with console frontend by @mpenn in NVIDIA#255
• [AIQ][25.05][RC3] Example to showcase Metadata support by @ericevans-nv in NVIDIA#256
• mem: If conversation is not provided build it from memory by @AnuradhaKaruppiah in NVIDIA#253
• Documentation restructure by @dagardner-nv in NVIDIA#189
• Prompt engineering to force ReAct agent to use memory for simple_rag example by @yczhang-nv in NVIDIA#260
• simple-calculator: Additional input validation by @AnuradhaKaruppiah in NVIDIA#259
• Removed simple_mcp example by @AnuradhaKaruppiah in NVIDIA#266
• Adding reference links to examples in README. by @ericevans-nv in NVIDIA#265
• mcp-client.md: Add a note to check that the MCP time-service is running by @AnuradhaKaruppiah in NVIDIA#267
• Remove username from README log by @yczhang-nv in NVIDIA#271
• Enhance error handling in MCP tool invocation by @mpenn in NVIDIA#263
• Resolves a linting error in MCP tool by @mpenn in NVIDIA#274
• Fix long-term memory issues of semantic_kernel example by @yczhang-nv in NVIDIA#270
• Update to reflect new naming guidelines by @dagardner-nv in NVIDIA#258
• Updating submodule that fixes UI broken links by @ericevans-nv in NVIDIA#273
• Change the example input for Multi Frameworks example by @yczhang-nv in NVIDIA#277
• Fix intermediate steps parents when the parent is a Tool by @mdemoret-nv in NVIDIA#269
• Set mcp-proxy version in the sample Dockerfile to 0.5 by @AnuradhaKaruppiah in NVIDIA#278
• Add an FAQ document by @dagardner-nv in NVIDIA#275
• Fix missing tool issue with profiler_agent example by @dagardner-nv in NVIDIA#279
• Add missing telemetry dependency to profiler_agent example by @dagardner-nv in NVIDIA#281
• eval-readme: Add instruction to copy the workflow output before re-runs by @AnuradhaKaruppiah in NVIDIA#280
• Add additional notes for intermittent long-term memory issues in examples by @yczhang-nv in NVIDIA#282
• Run tests on all supported versions of Python by @dagardner-nv in NVIDIA#283
• Fix the intermediate steps span logic to work better with nested coroutines and tasks by @mdemoret-nv in NVIDIA#285

New Contributors

• @dagardner-nv made their first contribution in NVIDIA#7
• @yczhang-nv made their first contribution in NVIDIA#9
• @gzitzlsb-nv made their first contribution in NVIDIA#25
• @atalhens made their first contribution in NVIDIA#28
• @avoroshilov made their first contribution in NVIDIA#42
• @balvisio made their first contribution in NVIDIA#22
• @ericevans-nv made their first contribution in NVIDIA#64
• @dnandakumar-nv made their first contribution in NVIDIA#70
• @wenqiglantz made their first contribution in NVIDIA#36
• @VictorYudin made their first contribution in NVIDIA#133
• @Hritik003 made their first contribution in NVIDIA#129
• @jkornblum-nv made their first contribution in NVIDIA#146
• @zac-wang-nv made their first contribution in NVIDIA#159
• @mpenn made their first contribution in NVIDIA#166
• @liamy-nv made their first contribution in NVIDIA#110
• @ayulockin made their first contribution in NVIDIA#170
• @soumilinandi made their first contribution in NVIDIA#191
• @hsin-c made their first contribution in NVIDIA#193

[1.1.0a1] - 2025-04-05

Added

• Added CORS configuration for the FastAPI server
• Added support for customizing evaluation outputs and uploading results to remote storage

Fixed

• Fixed aiq serve when running the simple_rag workflow example
• Added missing platformdirs dependency to aiqtoolkit package

[1.0.0] - 2024-12-04

Added

• First release.