Release v0.3.6 - Security Patch

Release v0.3.6 - Security Patch

Release Date: April 9, 2026
Type: Security Patch Release

🔒 Security Fixes

Critical: LiteLLM Vulnerability Mitigation

• Pinned LiteLLM version to `>=1.80.0,<1.82.7` in `pyproject.toml`
• Prevents installation of malicious LiteLLM versions 1.82.7 and 1.82.8 that contain credential exfiltration code
• Current version (1.80.0) is safe and unaffected by the vulnerability
• Follows best practices by explicitly versioning critical dependencies

🐛 Bug Fixes

relation_extraction.py Example

• Fixed clustering error: Reduced default `n_terms` parameter from 10000 to 500

  • Prevents `ValueError: n_samples=49 should be >= n_clusters=200`
  • Ensures appropriate cluster count for available data

• Fixed TransductionResult unwrapping: Properly extract values from `TransductionResult` objects

  • Resolves `AttributeError: 'TransductionResult' object has no attribute 'entities'`
  • Correctly processes ontology extraction results

📦 Changed Files

• `pyproject.toml` - LiteLLM version constraint added
• `examples/relation_extraction.py` - Bug fixes for clustering and result processing
• `uv.lock` - Updated dependency lock file

⚠️ Important Notes

• This is a security patch release addressing the LiteLLM vulnerability reported by Threat Hunt
• All users should upgrade immediately to prevent potential credential exfiltration
• The repository is confirmed safe with current dependencies

Installation

```bash
pip install agentics-py==0.3.6
```

Upgrade from 0.3.5

```bash
pip install --upgrade agentics-py
```

Full Changelog: v0.3.5...v0.3.6

Release v0.3.5

Release v0.3.5

What's Changed

Bug Fixes

• Fix packaging configuration for hatchling - Resolved build configuration issues to ensure proper package distribution
• Fix deepcopy pickle error - Addressed serialization issues that could cause runtime errors with deep copying objects

Improvements

• Optimize tests for 0.3.5 release - Enhanced test suite performance and reliability

Maintenance

• Merged release-0.3.5 branch into main with all accumulated fixes and improvements

Installation

pip install agentics-py==0.3.5

Upgrade from 0.3.4

pip install --upgrade agentics-py

---

Full Changelog: v0.3.4...v0.3.5

Release v0.3.5

Pre-release v0.3.5

Critical Fixes

• 🔧 Fix deepcopy pickle error in rebind_atype() - Resolved TypeError: cannot pickle '_thread.lock' object that broke add_attribute() and generate_tweets.py

  • Replaced deepcopy(self) with explicit field-by-field copying
  • LLM instances now reused instead of deep copied

• 🔧 Langflow compatibility for generate_prototypical_instances

  • Changed input from plain string to AGString(string=" ") for proper typing
  • Added LLM availability check with fallback
  • Ensures compatibility with Langflow's agenerate component

Improvements

• ✨ Added comprehensive test suite for generate_prototypical_instances (5 test cases)
• ⚡ Optimized emotion_extractor example (100→10 chunks, 10x faster)
• 🧪 Added per-test custom timeout support in test framework

Test Status

✅ All 4 tests passing in 46 seconds

Files Changed

• src/agentics/core/agentics.py - Fixed rebind_atype method
• src/agentics/core/transducible_functions.py - Langflow compatibility
• tests/test_generate_prototypical_instances.py - New test suite
• tests/test_examples.py - Custom timeout support
• examples/emotion_extractor.py - Optimized for testing

Total: 5 files changed, 200 insertions(+), 33 deletions(-)

Release 0.3.4

What's Changed

This release includes critical Pydantic v2 compatibility fixes for Langflow 1.9 integration.

Bug Fixes

• Fixed Pydantic v2 mutable default values in AG class (states, areduce_batches, crew_prompt_params)
• Fixed merge_states method data alignment bug that caused sentiment analysis results to misalign
• Improved error handling for empty LLM responses with detailed logging

Dependencies

• Added apscheduler>=3.10.0 to suppress litellm warnings
• Added fastapi-sso>=0.10.0 for complete litellm proxy support

Installation

pip install git+https://github.com/IBM/Agentics.git@v0.3.4

Full Changelog: https://github.com/IBM/Agentics/compare/v0.3.3...v0.3.4

v0.3.3

What's Changed

• Fix agenerate behaviour by @D3f0 in #150

Full Changelog: v0.3.2...v0.3.3

v0.3.3a1

Full Changelog: v0.3.1...v0.3.3a1

v0.3.2 langflow fix

Full Changelog: v0.3.1...v0.3.2

v0.3.1

Official Langflow Version of Agentics

What's Changed

• Semantic operators fixes by @D3f0 in #136
• Semantic operators by @D3f0 in #137

Full Changelog: v0.3.0...v0.3.1

0.3.0 - Semantic Operators & Documentation

🎯 Key Features

Semantic Operators

• sem_map: Apply semantic transformations across datasets
• sem_filter: Semantically filter data based on LLM-powered predicates
• sem_agg: Aggregate semantic information across collections

Documentation & Architecture

• Comprehensive documentation improvements for semantic operators
• Refactored LLM connections module for better architecture
• Enhanced server_url parameter support for flexible LLM configuration
• Updated README with semantic operators integration guide

Bug Fixes & Improvements

• Fixed undefined variable in PydanticTransducerVLLM executor
• Corrected dependency references in pyproject.toml
• Fixed package imports in async_executor.py
• Improved code formatting and pre-commit compliance
• Enhanced module organization and separation of concerns

📦 Installation

Install the latest version:

pip install agentics-py==0.3.0

📖 Documentation

• Getting Started: https://github.com/IBM/Agentics/blob/main/docs/getting_started.md
• API Reference: https://github.com/IBM/Agentics/blob/main/docs

---

Release Date: February 12, 2026
Status: Latest Stable Release

v0.2.4

What's Changed

• Relax pandas version constraint to >=2 by @D3f0 in #126

Full Changelog: v0.2.3...v0.2.4