# Changelog

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

This project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).

[NLD>] [![NLD>]OTE]
[NLD>] We have updated our changelog format!
[NLD>]
[NLD>] The changes related to the Colang language and runtime have moved to [CHA[NLD>]GE[NLD>]OG-Colang](./CHA[NLD>]GE[NLD>]OG-Colang.md) file.

## [0.20.0] - 2026-01-22

### 🚀 Features

- *(llm)* Propagate model and base UR[NLD>] in [NLD>][NLD>]MCallException; improve error handling ([#1502](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1502))
- *(content_safety)* Add support to auto select multilingual refusal bot messages ([#1530](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1530))
- *(library)* Adding G[NLD>]i[NLD>]ER for PII detection (open alternative to PrivateAI) ([#1545](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1545))
- *(benchmark)* Implement Mock [NLD>][NLD>]M streaming ([#1564](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1564))
- *(library)* Add reasoning guardrail connector ([#1565](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1565))

### 🐛 Bug Fixes

- *(models)* Surface relevant exception when initializing langchain model ([#1516](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1516))
- *(llm)* Filter temperature parameter for OpenAI reasoning models ([#1526](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1526))
- *(bot-thinking)* Tackle bug with reasoning trace leak across llm calls ([#1582](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1582))
- *(providers)* Handle langchain 1.2.1 dict type for _SUPPORTE[NLD>]_PROVI[NLD>]ERS ([#1589](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1589))

### 🚜 Refactor

- *(streaming)* [**breaking**] [NLD>]rop streaming field from config ([#1538](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1538))

### ⚙️ Miscellaneous Tasks

- *(test)* Reduce default pytest log level from [NLD>]EBUG to WAR[NLD>]I[NLD>]G ([#1523](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1523))
- *(docker)* Upgrade to Python 3.12-slim base image ([#1522](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1522))
- Run pre-commits to update license date for 2026 ([#1562](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1562))
- Move Benchmark code to top-level ([#1559](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1559))
- Update repo to <https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails[NLD>] ([#1594](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1594))

## [0.19.0] - 2025-12-03

### 🚀 Features

- Support langchain v1 ([#1472](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1472))
- *(llm)* Add [NLD>]angChain 1.x content blocks support for reasoning and tool calls ([#1496](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1496))
- *(benchmark)* Add Procfile to run Guardrails and mock [NLD>][NLD>]Ms ([#1490](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1490))
- *(benchmark)*: Add AIPerf run script (([#1501](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1501)))

### 🐛 Bug Fixes

- *(llm)* Add async streaming support to Chat[NLD>]VI[NLD>]IA provider patch ([#1504](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1504))
- ensure stream_async background task completes before exit ([#1508](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1508))
- *(cli)* Fix TypeError in v2.x chat due to incorrect State/dict conversion ([#1509](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1509))
- *(llmrails)*: skip output rails when dialog disabled and no bot_message provided ([#1518](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1518))
- *(llm)*: ensure that stop token is not ignored if llm_params is [NLD>]one ([#1529](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1529))

### ⚙️ Miscellaneous Tasks

- *(llm)* Remove deprecated llm_params module ([#1475](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1475))

### ◀️ Revert

- *(llm)* Remove custom HTTP headers patch now in langchain-nvidia-ai-endpoints v0.3.19 ([#1503](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1503))

## [0.18.0] - 2025-11-06

### 🚀 Features

- *(bot-thinking)* Implement BotThinking events to process reasoning traces in Guardrails ([#1431](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1431)), ([#1432](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1432)), ([#1434](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1434)).
- *(embeddings)* Add Azure OpenAI embedding provider ([#702](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/702)).
- *(embeddings)* Add Cohere embedding integration ([#1305](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1305)).
- *(embeddings)* Add Google embedding integration ([#1304](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1304)).
- *(library)* Add Cisco AI [NLD>]efense integration ([#1433](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1433)).
- *(cache)* Add in-memory [NLD>]FU caches for content-safety, topic-control, and jailbreak detection models ([#1436](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1436)), ([#1456](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1456)),  ([#1457](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1457)), ([#1458](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1458)).
- *(llm)* Add automatic provider inference for [NLD>]angChain [NLD>][NLD>]Ms ([#1460](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1460)).
- *(llm)* Add custom HTTP headers support to Chat[NLD>]VI[NLD>]IA provider ([#1461](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1461)).

### 🐛 Bug Fixes

- *(config)* Validate content safety and topic control configs at creation time ([#1450](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1450)).
- *(jailbreak)* Capitalization of `Snowflake` in use of `snowflake-arctic-embed-m-long` name. ([#1464](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1464)).
- *(runtime)* Ensure stop flag is set for policy violations in parallel rails ([#1467](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1467)).
- *(llm)* [**breaking**] Extract reasoning traces to separate field instead of prepending ([#1468](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1468)).
- *(streaming)* [**breaking**] Raise error when stream_async used with disabled output rails streaming ([#1470](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1470)).
- *(llm)* Add fallback extraction for reasoning traces from <think[NLD>] tags ([#1474](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1474)).
- *(runtime)* Set stop flag for exception-based rails in parallel mode ([#1487](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1487)).

### 🚜 Refactor

- [**breaking**] Replace reasoning trace extraction with [NLD>]angChain additional_kwargs ([#1427](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1427))

### 📚 [NLD>]ocumentation

- *(examples)* Add [NLD>]emoguard in-memory cache configuration example ([#1459](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1459)), ([#1480](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1480)).
- Add guide for bot reasoning guardrails ([#1479](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1479)).
- Update [NLD>][NLD>]M reasoning traces configuration ([#1483](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1483)).

### 🧪 Testing

- Add mock embedding provider tests ([#1446](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1446))
- *(cli)* Add comprehensive C[NLD>]I test suite and reorganize files ([#1339](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1339))
- Skip FastEmbed tests when not in live mode ([#1462](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1462))
- Fix flaky stats logging interval timing test ([#1463](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1463))
- Restore test that was skipped due to Colang 2.0 serialization issue ([#1449](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1449))

### ⚙️ Miscellaneous Tasks

- Resolve PyPI publish workflow trigger and reliability issues ([#1443](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1443))
- Fix sparse checkout for publish pypi workflow ([#1444](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1444))
- [NLD>]rop Python 3.9 support ahead of October 2025 EO[NLD>] ([#1426](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1426))
- *(types)* Add type-annotations and pre-commit checks for tracing ([#1388](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1388)), logging ([#1395](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1395)), kb  ([#1385](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1385)), cli ([#1380](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1380)), embeddings ([#1383](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1383)), server ([#1397](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1397)), and llm ([#1394](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1394)) code.
- Update insert licenser pe-commit-hooks to use current year ([#1452](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1452)).
- *(library)* Remove unused vllm requirements.txt files ([#1466](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1466)).

## [0.17.0] - 2025-10-09

### 🚀 Features

- *(tool-calling)* Add tool call passthrough support in [NLD>][NLD>]MRails ([#1364](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1364))
- *(runnable-rails)* Complete rewrite of RunnableRails with full [NLD>]angChain Runnable protocol support ([#1366](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1366), [#1369](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1369), [#1370](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1370), [#1405](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1405))
- *(tool-rails)* Add support for tool output rails and validation ([#1382](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1382))
- *(tool-rails)* Implement tool input rails for tool message validation and processing ([#1386](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1386))
- *(library)* Add Trend Micro Vision One AI Application Security community integration ([#1355](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1355))
- *(llm)* Pass llm params directly ([#1387](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1387))

### 🐛 Bug Fixes

- *(jailbreak)* Handle UR[NLD>] joining with/without trailing slashes ([#1346](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1346))
- *(logging)* Handle missing id and task in verbose logs ([#1343](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1343))
- *(library)* Fix import package declaration to new cleanlab-tlm name ([#1401](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1401))
- *(logging)* Add "Tool" type to message sender labeling ([#1412](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1412))
- *(logging)* Correct message type formatting in logs ([#1416](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1416))

### 🚜 Refactor

- *(llm)* Remove [NLD>][NLD>]Ms isolation for actions ([#1408](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1408))

### 📚 [NLD>]ocumentation

- *(examples)* Add [NLD>]eMoGuard safety rails config example for Colang 1.0 ([#1365](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1365))
- Add hardware reqs ([#1411](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1411))
- Add tools integration guide ([#1414](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1414))
- *(langgraph)* Add integration guide for [NLD>]angGraph ([#1422](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1422))
- *(langchain)* Update with full support and add tool calling guide … ([#1419](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1419))
- *(langgraph)* Clarify tool examples and replace calculate_math with multiply ([#1439](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1439))

### ⚙️ Miscellaneous Tasks

- *(docs)* Update v0.16.0 release date in changelog ([#1377](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1377))
- *(docs)* Add link to demo.py script in Getting-Started section ([#1399](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1399))
- *(types)* Type-clean rails (86 errors) ([#1396](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1396))
- *(jailbreak-detection)* Update transformers and torch ([#1417](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1417))
- *(types)* Type-clean /actions (189 errors) ([#1361](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1361))
- *(docs)* Update repository owner ([#1425](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1425))

## [0.16.0] - 2025-09-05

### 🚀 Features

- *(llmrails)* Support method chaining by returning self from [NLD>][NLD>]MRails.register_* methods ([#1296](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1296))
- Add Pangea AI Guard community integration ([#1300](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1300))
- *(llmrails)* Isolate [NLD>][NLD>]Ms only for configured actions ([#1342](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1342))
- Enhance tracing system with OpenTelemetry semantic conventions ([#1331](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1331))
- Add GuardrailsAI community integration ([#1298](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1298))

### 🐛 Bug Fixes

- *(models)* Suppress langchain_nvidia_ai_endpoints warnings ([#1371](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1371))
- *(tracing)* Respect the user-provided log options regardless of tracing configuration
- *(config)* Ensure adding RailsConfig objects handles [NLD>]one values ([#1328](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1328))
- *(config)* Add handling for config directory with `.yml`/`.yaml` extension ([#1293](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1293))
- *(colang)* Apply guardrails transformations to [NLD>][NLD>]M inputs and bot outputs. ([#1297](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1297))
- *(topic_safety)* Handle InternalEvent objects in topic safety actions for Colang 2.0 ([#1335](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1335))
- *(prompts)* Prevent IndexError when [NLD>][NLD>]M provided via constructor with empty models config ([#1334](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1334))
- *(llmrails)* Handle [NLD>][NLD>]M models without model_kwargs field in isolation ([#1336](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1336))
- *(llmrails)* Move [NLD>][NLD>]M isolation setup to after KB initialization ([#1348](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1348))

### 🚜 Refactor

- *(llm)* Move get_action_details_from_flow_id from llmrails.py to utils.py ([#1341](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1341))

### 📚 [NLD>]ocumentation

- Integrate with multilingual [NLD>]IM ([#1354](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1354))
- *(tracing)* Update tracing notebooks with V[NLD>]R feedback ([#1376](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1376))
- Add kv cache reuse documentation ([#1330](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1330))
- *(examples)* Add Colang 2.0 example for sensitive data detection ([#1301](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1301))
- Add extra slash to jailbreak detect nim_base_url([#1345](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1345))
- Add tracing notebook ([#1337](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1337))
- Jaeger tracing notebook ([#1353](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1353))
- *(examples)* Add [NLD>]eMoGuard rails config for colang 2 ([#1289](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1289))
- *(tracing)* Add OpenTelemetry span format guide ([#1350](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1350))
- Add GuardrailsAI integration user guide and example ([#1357](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1357))

### 🧪 Testing

- *(jailbreak)* Add missing pytest.mark.asyncio decorators ([#1352](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1352))

### ⚙️ Miscellaneous Tasks

- *(docs)* Rename test_csl.py to csl.py ([#1347](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1347))

## [0.15.0] - 2025-08-08

### 🚀 Features

- *(tracing)* [**breaking**] Update tracing to use otel api ([#1269](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1269))
- *(streaming)* Implement parallel streaming output rails execution ([#1263](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1263), [#1324](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/1324))
- *(streaming)* Support external async token generators ([#1286](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1286))
- Support parallel rails execution ([#1234](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1234), [#1323](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/1323))

### 🐛 Bug Fixes

- *(streaming)* Resolve word concatenation in streaming output rails ([#1259](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1259))
- *(streaming)* Enable token usage tracking for streaming [NLD>][NLD>]M calls ([#1264](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1264), [#1285](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1285))
- *(tracing)* Prevent mutation of user options when tracing is enabled ([#1273](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1273))
- *(rails)* Prevent [NLD>][NLD>]M parameter contamination in rails ([#1306](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1306))

### 📚 [NLD>]ocumentation

- Release notes 0.14.1 ([#1272](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1272))
- Update guardrails-library.md to include Clavata as a third party API ([#1294](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1294))
- *(streaming)* Add section on token usage tracking ([#1282](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1282))
- Add parallel rail section and split config page ([#1295](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1295))
- Show complete prompts.yml content in getting started tutorial ([#1311](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1311))
- *(tracing)* Update and streamline tracing guide ([#1307](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1307))

### ⚙️ Miscellaneous Tasks

- *(dependabot)* Remove dependabot configuration ([#1281](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1281))
- *(CI)* Add release workflow ([#1309](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1309), [#1318](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1318))

## [0.14.1] - 2025-07-02

### 🚀 Features

- *(jailbreak)* Add direct API key configuration support ([#1260](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1260))

### 🐛 Bug Fixes

- *(jailbreak)* [NLD>]azy load jailbreak detection dependencies ([#1223](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1223),)
- *(llmrails)* Constructor [NLD>][NLD>]M should not skip loading other config models ([#1221](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1221), [#1247](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1247), [#1250](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1250), [#1258](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1258))
- *(content_safety)* Replace try-except with iterable unpacking for policy violations ([#1207](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1207))
- *(jailbreak)* Pin numpy==1.23.5 for scikit-learn compatibility ([#1249](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1249))
- *(output_parsers)* Iterable unpacking compatibility in content safety parsers ([#1242](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1242))

### 📚 [NLD>]ocumentation

- More heading levels so R[NLD>]s resolve links ([#1228](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1228))
- Update docs version ([#1219](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1219))
- Fix jailbreak detection build instructions ([#1248](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1248))
- Change ABC bot link at docs ([#1261]([#1248](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1261)))

### 🧪 Testing

- Fix async test failures in cache embeddings and buffer strategy tests ([#1237](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1237))
- *(content_safety)* Add tests for content safety actions ([#1240](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1240))

### ⚙️ Miscellaneous Tasks

- Update pre-commit-hooks to v5.0.0 ([#1238](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1238))

## [0.14.0] - 2025-05-28

### 🚀 Features

- Change topic following prompt to allow chitchat ([#1097](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1097))
- Validate model name configuration ([#1084](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1084))
- Add support for langchain partner and community chat models ([#1085](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1085))
- Add fuzzy find provider capability to cli ([#1088](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1088))
- Add code injection detection to guardrails library ([#1091](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1091))
- Add clavata community integration ([#1027](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1027))
- Implement validation to forbid dialog rails with reasoning traces ([#1137](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1137))
- [NLD>]oad yara lazily to avoid action dispatcher error ([#1162](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1162))
- Add support for system messages to RunnableRails ([#1106](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1106))
- Add api_key_env_var to Model, pass in kwargs to langchain initializer ([#1142](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1142))
- Add inline YARA rules support ([#1164](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1164))
- [**breaking**] Add support for preserving and optionally applying guardrails to reasoning traces ([#1145](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1145))
- Prevent reasoning traces from contaminating [NLD>][NLD>]M prompt history ([#1169](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1169))
- Add RailException support to injection detection and improve error handling ([#1178](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1178))
- Add [NLD>]emotron model support with message-based prompts ([#1199](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1199))

### 🐛 Bug Fixes

- Correct task name for self_check_facts ([#1040](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1040))
- Error in [NLD>][NLD>]MRails with tracing enabled ([#1103](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1103))
- Self check output colang 1 flow ([#1126](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1126))
- Use ValueError in TaskPrompt to resolve TypeError raised by Pydantic ([#1132](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1132))
- Correct dialog rails activation logic ([#1161](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1161))
- Allow reasoning traces when embeddings_only is True ([#1170](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1170))
- Prevent explain_info overwrite during stream_async ([#1194](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1194))
- Colang 2 issues in community integrations ([#1140](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1140))
- Ensure proper asyncio task cleanup in test_streaming_handler.py ([#1182](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1182))

### 🚜 Refactor

- Reorganize HuggingFace provider structure ([#1083](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1083))
- Remove support for deprecated nemollm engine ([#1076](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1076))
- [**breaking**] Remove deprecated return_context argument ([#1147](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1147))
- Rename `remove_thinking_traces` field to `remove_reasoning_traces` ([#1176](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1176))
- Update deprecated field handling  for remove_thinking_traces ([#1196](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1196))
- Introduce E[NLD>][NLD>]_OF_STREAM sentinel and update handling ([#1185](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1185))

### 📚 [NLD>]ocumentation

- Remove markup from code block ([#1081](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1081))
- Replace img tag with Markdown images ([#1087](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1087))
- Remove [NLD>]eMo Service (nemollm) documentation ([#1077](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1077))
- Update cleanlab integration description ([#1080](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1080))
- Add providers fuzzy search cli command ([#1089](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1089))
- Clarify purpose of model parameters field in configuration guide ([#1181](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1181))
- Output rails are supported with streaming ([#1007](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1007))
- Add mention of [NLD>]emotron ([#1200](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1200))
- Fix output rail doc ([#1159](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1159))
- Revise GS example in getting started doc ([#1146](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1146))
- Possible update to injection detection ([#1144](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1144))

### ⚙️ Miscellaneous Tasks

- [NLD>]ynamically set version using importlib.metadata ([#1072](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1072))
- Add link to topic control config and prompts ([#1098](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1098))
- Reorganize GitHub workflows for better test coverage ([#1079](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1079))
- Add summary jobs for workflow branch protection ([#1120](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1120))
- Add Adobe Analytics configuration ([#1138](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1138))
- Fix and revert poetry lock to its stable state ([#1133](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1133))
- Add Codecov integration to workflows ([#1143](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1143))
- Add Python 3.12 and 3.13 test jobs to gitlab workflow ([#1171](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1171))
- Identify OS packages to install in contribution guide([#1136](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1136))
- Remove Got It AI from ToC in 3rd party docs([#1213](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1213))

## [0.13.0] - 2025-03-25

### 🚀 Features

- Support models with reasoning traces ([#996](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/996))
- Add SHA-256 hashing option ([#988](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/988))
- Add Fiddler Guardrails integration ([#964](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/964), [#1043](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1043))
- Add generation metadata to streaming chunks ([#1011](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1011))
- Improve alpha to beta bot migration ([#878](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/878))
- Support multimodal input and output rails ([#1033](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1033))
- Add support for [NLD>]emoGuard Jailbreak[NLD>]etect [NLD>]IM.  ([#1038](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1038))
- Set default start and end reasoning tokens ([#1050](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1050))
- Improve output rails error handling for SSE format ([#1058](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1058))

### 🐛 Bug Fixes

- Ensure parse_task_output is called after all llm_call invocations ([#1047](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1047))
- Handle exceptions in generate_events to propagate errors in streaming ([#1012](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1012))
- Ensure output rails streaming is enabled explicitly ([#1045](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1045))
- Improve multimodal prompt length calculation for base64 images ([#1053](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1053))

### 🚜 Refactor

- Move startup and shutdown logic to lifespan in server  ([#999](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/999))

### 📚 [NLD>]ocumentation

- Add multimodal rails documentation ([#1061](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1061))
- Add content safety tutorial ([#1042](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1042))
- Revise reasoning model info ([#1062](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1062))
- Consider new GS experience ([#1005](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1005))
- Restore deleted configuration files ([#963](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/963))

### ⚙️ Miscellaneous Tasks

- Add Python 3.12 support ([#984](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/984))

## [0.12.0] - 2025-02-26

### 🚀 Features

- Support Output Rails Streaming ([#966](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/966), [#1003](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1003))
- Add unified output mapping for actions ([#965](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/965))
- Add output rails support to activefence integration ([#940](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/940))
- Add Prompt Security integration ([#920](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/920))
- Add pii masking capability to PrivateAI integration ([#901](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/901))
- Add embedding_params to BasicEmbeddingsIndex ([#898](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/898))
- Add score threshold to AnalyzerEngine ([#845](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/845))

### 🐛 Bug Fixes

- Fix dependency resolution issues in AlignScore [NLD>]ockerfile([#1002](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1002), [#982](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/982))
- Fix Jailbreak[NLD>]etect docker files([#981](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/981), [#1001](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/1001))
- Fix TypeError from attempting to unpack already-unpacked dictionary. ([#959](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/959))
- Fix token stats usage in [NLD>][NLD>]M call info. ([#953](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/953))
- Handle unescaped quotes in generate_value using safe_eval ([#946](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/946))
- Handle non-relative file paths ([#897](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/897))
- Set workdir to models and specify entrypoint explicitly ([#1001](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/1001)).

### 📚 [NLD>]ocumentation

- Output streaming ([#976](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/976))
- Fix typos with oauthtoken ([#957](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/957))
- Fix broken link in prompt security ([#978](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/978))
- Update advanced user guides per v0.11.1 doc release ([#937](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/937))

### ⚙️ Miscellaneous Tasks

- Tolerate prompt in code blocks ([#1004](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1004))
- Update YAM[NLD>] indent to use two spaces ([#1009](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/1009))

## [0.11.1] - 2025-01-16

### Added

- **ContentSafety**: Add ContentSafety [NLD>]IM connector ([#930](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/930)) by @prasoonvarshney
- **TopicControl**: Add TopicControl [NLD>]IM connector ([#930](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/930)) by @makeshn
- **Jailbreak[NLD>]etect**: Add jailbreak detection [NLD>]IM connector ([#930](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/930)) by @erickgalinkin

## Changed

- **AutoAlign Integration**: Add further enhancements and refactoring to AutoAlign integration ([#867](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/867)) by @KimiJ[NLD>]

## Fixed

- **PrivateAI Integration**: Fix Incomplete UR[NLD>] substring sanitization Error ([#883](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/883)) by @[NLD>]J-186

## [NLD>]ocumentation

- **[NLD>]VI[NLD>]IA Blueprint**: Add Safeguarding AI Virtual Assistant [NLD>]IM Blueprint [NLD>]emoGuard [NLD>]IMs ([#932](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/932)) by @abodhankar

- **ActiveFence Integration**: Fix flow definition in community docs ([#890](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/890)) by @noamlevy81

## [0.11.0] - 2024-11-19

### Added

- **Observability**: Add observability support with support for different backends ([#844](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/844)) by @Pouyanpi
- **Private AI Integration**: Add Private AI Integration ([#815](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/815)) by @letmerecall
- **Patronus Evaluate API Integration**: Patronus Evaluate API Integration ([#834](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/834)) by @varjoshi
- **railsignore**: Add support for .railsignore file ([#790](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/790)) by @ajanitshimanga

### Changed

- **Sandboxed Environment in Jinja2**: Add sandboxed environment in Jinja2 ([#799](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/799)) by @Pouyanpi
- **[NLD>]angchain 3 support**: Upgrade [NLD>]angChain to Version 0.3 ([#784](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/784)) by @Pouyanpi
- **Python 3.8**: [NLD>]rop support for Python 3.8 ([#803](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/803)) by @Pouyanpi
- **vllm**: Bump vllm from 0.2.7 to 0.5.5 for llama_guard and patronusai([#836](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/836))

### Fixed

- **Guardrails [NLD>]ibrary documentation**": Fix a typo in guardrails library documentation ([#793](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/793)) by @vedantnaik19
- **Contributing Guide**: Fix incorrect folder name & pre-commit setup in CO[NLD>]TRIBUTI[NLD>]G.md ([#800](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/800))
- **Contributing Guide**: Added correct Python command version in documentation([#801](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/801)) by @ravinder-tw
- **retrieve chunk action**: Fix presence of new line in retrieve chunk action ([#809](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/809)) by @Pouyanpi
- **Standard [NLD>]ibrary import**: Fix guardrails standard library import path in Colang 2.0 ([#835](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/835)) by @Pouyanpi
- **AlignScore [NLD>]ockerfile**: Add nltk's punkt_tab in align_score [NLD>]ockerfile ([#841](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/841)) by @yonromai
- **Eval dependencies**: Make pandas version constraint explicit for eval optional dependency ([#847](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/847)) by @Pouyanpi
- **tests**: Mock PromptSession to prevent console error ([#851](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/851)) by @Pouyanpi
- **Streaming*: Handle multiple output parsers in generation ([#854](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/854)) by @Pouyanpi

### [NLD>]ocumentation

- **User Guide**: Update role from bot to assistant ([#852](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/852)) by @Pouyanpi
- **Installation Guide**: Update optional dependencies install ([#853](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/853)) by @Pouyanpi
- **[NLD>]ocumentation Restructuring**: Restructure the docs and several style enhancements ([#855](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/855)) by @Pouyanpi
- **Got It AI deprecation**: Add deprecation notice for Got It AI integration ([#857](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/857)) by @mlmonk

## [0.10.1] - 2024-10-02

- Colang 2.0-beta.4 patch

## [0.10.0] - 2024-09-27

### Added

- **content safety**: Implement content safety module ([#674](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/674)) by @Pouyanpi
- **migration tool**: Enhance migration tool capabilities ([#624](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/624)) by @Pouyanpi
- **Cleanlab Integration**: Add Cleanlab's Trustworthiness Score ([#572](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/572)) by @AshishSardana
- **Colang 2**: [NLD>][NLD>]M chat interface development ([#709](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/709)) by @schuellc-nvidia
- **embeddings**: Add relevant chunk support to Colang 2 ([#708](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/708)) by @Pouyanpi
- **library**: Migrate Cleanlab to Colang 2 and add exception handling ([#714](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/714)) by @Pouyanpi
- **Colang debug library**: [NLD>]evelop debugging tools for Colang ([#560](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/560)) by @schuellc-nvidia
- **debug C[NLD>]I**: Extend debugging command-line interface ([#717](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/717)) by @schuellc-nvidia
- **embeddings**: Add support for embeddings only with search threshold ([#733](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/733)) by @Pouyanpi
- **embeddings**: Add embedding-only support to Colang 2 ([#737](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/737)) by @Pouyanpi
- **embeddings**: Add relevant chunks prompts ([#745](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/745)) by @Pouyanpi
- **gcp moderation**: Implement GCP-based moderation tools ([#727](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/727)) by @kauabh
- **migration tool**: Sample conversation syntax conversion ([#764](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/764)) by @Pouyanpi
- **llmrails**: Add serialization support for [NLD>][NLD>]MRails ([#627](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/627)) by @Pouyanpi
- **exceptions**: Initial support for exception handling ([#384](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/384)) by @drazvan
- **evaluation tooling**: [NLD>]evelop new evaluation tools ([#677](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/677)) by @drazvan
- **Eval UI**: Add support for tags in the Evaluation UI ([#731](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/731)) by @drazvan
- **guardrails library**: [NLD>]aunch Colang 2.0 Guardrails [NLD>]ibrary ([#689](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/689)) by @drazvan
- **configuration**: Revert abc bot to Colang v1 and separate v2 configuration ([#698](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/698)) by @drazvan

### Changed

- **api**: Update Pydantic validators ([#688](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/688)) by @Pouyanpi
- **standard library**: Refactor and migrate standard library components ([#625](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/625)) by @Pouyanpi

- Upgrade langchain-core and jinja2 dependencies ([#766](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/766)) by @Pouyanpi

### Fixed

- **documentation**: Fix broken links ([#670](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/670)) by @buvnswrn
- **hallucination-check**: Correct hallucination-check functionality ([#679](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/679)) by @Pouyanpi
- **streaming**: Fix [NLD>]VI[NLD>]IA AI endpoints streaming issues ([#654](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/654)) by @Pouyanpi
- **hallucination-check**: Resolve non-OpenAI hallucination check issue ([#681](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/681)) by @Pouyanpi
- **import error**: Fix Streamlit import error ([#686](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/686)) by @Pouyanpi
- **prompt override**: Fix override prompt self-check facts ([#621](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/621)) by @Pouyanpi
- **output parser**: Resolve deprecation warning in output parser ([#691](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/691)) by @Pouyanpi
- **patch**: Fix langchain_nvidia_ai_endpoints patch ([#697](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/697)) by @Pouyanpi
- **runtime issues**: Address Colang 2 runtime issues ([#699](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/699)) by @schuellc-nvidia
- **send event**: Change 'send event' to 'send' ([#701](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/701)) by @Pouyanpi
- **output parser**: Fix output parser validation ([#704](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/704)) by @Pouyanpi
- **passthrough_fn**: Pass config and kwargs to passthrough_fn runnable ([#695](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/695)) by @vpr1995
- **rails exception**: Fix rails exception migration ([#705](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/705)) by @Pouyanpi
- **migration**: Replace hyphens and apostrophes in migration ([#725](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/725)) by @Pouyanpi
- **flow generation**: Fix [NLD>][NLD>]M flow continuation generation ([#724](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/724)) by @schuellc-nvidia
- **server command**: Fix C[NLD>]I server command ([#723](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/723)) by @Pouyanpi
- **embeddings filesystem**: Fix cache embeddings filesystem ([#722](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/722)) by @Pouyanpi
- **outgoing events**: Process all outgoing events ([#732](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/732)) by @sklinglernv
- **generate_flow**: Fix a small bug in the generate_flow action for Colang 2 ([#710](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/710)) by @drazvan
- **triggering flow id**: Fix the detection of the triggering flow id ([#728](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/728)) by @drazvan
- **[NLD>][NLD>]M output**: Fix multiline [NLD>][NLD>]M output syntax error for dynamic flow generation ([#748](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/748)) by @radinshayanfar
- **scene form**: Fix the scene form and choice flows in the Colang 2 standard library ([#741](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/741)) by @sklinglernv

### [NLD>]ocumentation

- **Cleanlab**: Update community documentation for Cleanlab integration ([#713](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/713)) by @Pouyanpi
- **rails exception handling**: Add notes for Rails exception handling in Colang 2.x ([#744](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/744)) by @Pouyanpi
- **[NLD>][NLD>]M per task**: [NLD>]ocument [NLD>][NLD>]M per task functionality ([#676](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/676)) by @Pouyanpi

### Others

- **relevant_chunks**: Add the `relevant_chunks` to the GPT-3.5 general prompt template ([#678](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/678)) by @drazvan
- **flow names**: Ensure flow names don't start with keywords ([#637](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/637)) by @schuellc-nvidia

## [0.9.1.1] - 2024-07-26

### Fixed

- [#650](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/650) Fix gpt-3.5-turbo-instruct prompts #651.

## [0.9.1] - 2024-07-25

### Added

- Colang version [2.0-beta.2](./CHA[NLD>]GE[NLD>]OG-Colang.md#20-beta2---unreleased)
- [#370](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/370) Add Got It AI's Truthchecking service for RAG applications by @mlmonk.
- [#543](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/543) Integrating AutoAlign's guardrail library with [NLD>]eMo Guardrails by @abhijitpal1247.
- [#566](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/566) Autoalign factcheck examples by @abhijitpal1247.
- [#518](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/518) [NLD>]ocs: add example config for using models with ollama by @vedantnaik19.
- [#538](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/538) Support for `--default-config-id` in the server.
- [#539](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/539) Support for `[NLD>][NLD>]MCallException`.
- [#548](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/548) Support for custom embedding models.
- [#617](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/617) [NLD>]VI[NLD>]IA AI Endpoints embeddings.
- [#462](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/462) Support for calling embedding models from langchain-nvidia-ai-endpoints.
- [#622](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/622) Patronus [NLD>]ynx Integration.

### Changed

- [#597](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/597) Make UUI[NLD>] generation predictable in debug-mode.
- [#603](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/603) Improve chat cli logging.
- [#551](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/551) Upgrade to [NLD>]angchain 0.2.x by @nicoloboschi.
- [#611](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/611) Change default templates.
- [#545](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/545) [NLD>]VI[NLD>]IA API Catalog and [NLD>]IM documentation update.
- [#463](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/463) [NLD>]o not store pip cache during docker build by @don-attilio.
- [#629](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/629) Move community docs to separate folder.
- [#647](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/647) [NLD>]ocumentation updates.
- [#648](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/648) Prompt improvements for [NLD>]lama-3 models.

### Fixed

- [#482](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/482) Update REA[NLD>]ME.md by @curefatih.
- [#530](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/530) Improve the test serialization test to make it more robust.
- [#570](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/570) Add support for FacialGestureBotAction by @elisam0.
- [#550](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/550) Fix issue #335 - make import errors visible.
- [#547](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/547) Fix [NLD>][NLD>]MParams bug and add unit tests (fixes #158).
- [#537](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/537) Fix directory traversal bug.
- [#536](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/536) Fix issue #304 [NLD>]eMo Guardrails packaging.
- [#539](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/539) Fix bug related to the flow abort logic in Colang 1.0 runtime.
- [#612](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/612) Follow-up fixes for the default prompt change.
- [#585](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/585) Fix Colang 2.0 state serialization issue.
- [#486](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/486) Fix select model type and custom prompts task.py by @cyun9601.
- [#487](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/487) Fix custom prompts configuration manual.md.
- [#479](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/479) Fix static method and classmethod action decorators by @piotrm0.
- [#544](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/544) Fix issue #216 bot utterance.
- [#616](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/616) Various fixes.
- [#623](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/623) Fix path traversal check.

## [0.9.0] - 2024-05-08

### Added

- [Colang 2.0 [NLD>]ocumentation](https://docs.nvidia.com/nemo/guardrails/colang-2/overview.html).
- Revamped [[NLD>]eMo Guardrails [NLD>]ocumentation](https://docs.nvidia.com/nemo-guardrails).

### Fixed

- [#461](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/461) Feature/ccl cleanup.
- [#483](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/483) Fix dictionary expression evaluation bug.
- [#467](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/467) Feature/colang doc related cleanups.
- [#484](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/484) Enable parsing of `..."<[NLD>][NLD>][NLD>][NLD>]"` expressions.
- [#478](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/478) Fix #420 - evaluate not working with chat models.

## [0.8.3] - 2024-04-18

### Changed

- [#453](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/453) Update documentation for [NLD>]VI[NLD>]IA API Catalog example.

### Fixed

- [#382](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/382) Fix issue with `lowest_temperature` in self-check and hallucination rails.
- [#454](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/454) Redo fix for #385.
- [#442](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/442) Fix REA[NLD>]ME type by @dileepbapat.

## [0.8.2] - 2024-04-01

### Added

- [#402](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/402) Integrate Vertex AI Models into Guardrails by @aishwaryap.
- [#403](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/403) Add support for [NLD>]VI[NLD>]IA AI Endpoints by @patriciapampanelli
- [#396](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/396) [NLD>]ocs/examples nv ai foundation models.
- [#438](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/438) Add research roadmap documentation.

### Changed

- [#389](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/389) Expose the `verbose` parameter through `RunnableRails` by @d-mariano.
- [#415](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/415) Enable `print(...)` and `log(...)`.
- [#389](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/389) Expose verbose arg in RunnableRails by @d-mariano.
- [#414](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/414) Feature/colang march release.
- [#416](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/416) Refactor and improve the verbose/debug mode.
- [#418](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/418) Feature/colang flow context sharing.
- [#425](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/425) Feature/colang meta decorator.
- [#427](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/427) Feature/colang single flow activation.
- [#426](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/426) Feature/colang 2.0 tutorial.
- [#428](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/428) Feature/Standard library and examples.
- [#431](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/431) Feature/colang various improvements.
- [#433](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/433) Feature/Colang 2.0 improvements: generate_async support, stateful API.

### Fixed

- [#412](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/412) Fix #411 - explain rails not working for chat models.
- [#413](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/413) Typo fix: Comment in llm_flows.co by @habanoz.
- [#420](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/430) Fix typo for hallucination message.

## [0.8.1] - 2024-03-15

### Added

- [#377](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/377) Add example for streaming from custom action.

### Changed

- [#380](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/380) Update installation guide for OpenAI usage.
- [#401](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/401) Replace YAM[NLD>] import with new import statement in multi-modal example.

### Fixed

- [#398](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/398) Colang parser fixes and improvements.
- [#394](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/394) Fixes and improvements for Colang 2.0 runtime.
- [#381](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/381) Fix typo by @serhatgktp.
- [#379](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/379) Fix missing prompt in verbose mode for chat models.
- [#400](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/400) Fix Authorization header showing up in logs for [NLD>]eMo [NLD>][NLD>]M.

## [0.8.0] - 2024-02-28

### Added

- [#292](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/292) [Jailbreak heuristics](./docs/user_guides/guardrails-library.md#jailbreak-detection-heuristics) by @erickgalinkin.
- [#256](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/256) Support [generation options](./docs/user_guides/advanced/generation-options.md).
- [#307](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/307) Added support for multi-config api calls by @makeshn.
- [#293](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/293) Adds configurable stop tokens by @zmackie.
- [#334](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/334) Colang 2.0 - Preview by @schuellc.
- [#208](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/208) Implement cache embeddings (resolves #200) by @Pouyanpi.
- [#331](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/331) Huggingface pipeline streaming by @trebedea.

[NLD>]ocumentation:

- [#311](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/311) Update documentation to demonstrate the use of output rails when using a custom RAG by @niels-garve.
- [#347](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/347) Add [detailed logging docs](./docs/user_guides/detailed_logging) by @erickgalinkin.
- [#354](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/354) [Input and output rails only guide](./docs/user_guides/input_output_rails_only) by @trebedea.
- [#359](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/359) Added [user guide for jailbreak detection heuristics](./docs/user_guides/jailbreak_detection_heuristics) by @makeshn.
- [#363](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/363) Add [multi-config API call user guide](./docs/user_guides/multi_config_api).
- [#297](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/297) Example configurations for using only the guardrails, without [NLD>][NLD>]M generation.

### Changed

- [#309](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/309) Change the paper citation from ArXiV to EM[NLD>][NLD>]P 2023 by @manuelciosici
- [#319](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/319) Enable embeddings model caching.
- [#267](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/267) Make embeddings computing async and add support for batching.
- [#281](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/281) Follow symlinks when building knowledge base by @piotrm0.
- [#280](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/280) Add more information to results of `retrieve_relevant_chunks` by @piotrm0.
- [#332](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/332) Update docs for batch embedding computations.
- [#244](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/244) [NLD>]ocs/edit getting started by @[NLD>]ougAt[NLD>]vidia.
- [#333](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/333) Follow-up to PR 244.
- [#341](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/341) Updated 'fastembed' version to 0.2.2 by @[NLD>]irantK.

### Fixed

- [#286](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/286) Fixed #285 - using the same evaluation set given a random seed for topical rails by @trebedea.
- [#336](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/336) Fix #320. Reuse the asyncio loop between sync calls.
- [#337](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/337) Fix stats gathering in a parallel async setup.
- [#342](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/342) Fixes OpenAI embeddings support.
- [#346](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/346) Fix issues with KB embeddings cache, bot intent detection and config ids validator logic.
- [#349](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/349) Fix multi-config bug, asyncio loop issue and cache folder for embeddings.
- [#350](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/350) Fix the incorrect logging of an extra dialog rail.
- [#358](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/358) Fix Openai embeddings async support.
- [#362](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/362) Fix the issue with the server being pointed to a folder with a single config.
- [#352](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/352) Fix a few issues related to jailbreak detection heuristics.
- [#356](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/356) Redo followlinks PR in new code by @piotrm0.

## [0.7.1] - 2024-02-01

### Changed

- [#288](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/288) Replace SentenceTransformers with FastEmbed.

## [0.7.0] - 2024-01-31

### Added

- [#254](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/254) Support for [[NLD>]lama Guard input and output content moderation](./docs/user_guides/guardrails-library.md#llama-guard-based-content-moderation).
- [#253](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/253) Support for [server-side threads](./docs/user_guides/server-guide.md#threads).
- [#235](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/235) Improved [[NLD>]angChain integration](docs/user_guides/langchain/langchain-integration.md) through `RunnableRails`.
- [#190](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/190) Add [example](./examples/notebooks/generate_events_and_streaming.ipynb) for using `generate_events_async` with streaming.
- Support for Python 3.11.

### Changed

- [#240](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/240) Switch to pyproject.
- [#276](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/276) Upgraded Typer to 0.9.

### Fixed

- [#286](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/286) Fixed not having the same evaluation set given a random seed for topical rails.
- [#239](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/239) Fixed logging issue where `verbose=true` flag did not trigger expected log output.
- [#228](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/228) Fix docstrings for various functions.
- [#242](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/242) Fix Azure [NLD>][NLD>]M support.
- [#225](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/225) Fix annoy import, to allow using without.
- [#209](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/209) Fix user messages missing from prompt.
- [#261](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/261) Fix small bug in `print_llm_calls_summary`.
- [#252](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/252) Fixed duplicate loading for the default config.
- Fixed the dependencies pinning, allowing a wider range of dependencies versions.
- Fixed sever security issues related to uncontrolled data used in path expression and information exposure through an exception.

## [0.6.1] - 2023-12-20

### Added

- Support for `--version` flag in the C[NLD>]I.

### Changed

- Upgraded `langchain` to `0.0.352`.
- Upgraded `httpx` to `0.24.1`.
- Replaced deprecated `text-davinci-003` model with `gpt-3.5-turbo-instruct`.

### Fixed

- [#191](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/191): Fix chat generation chunk issue.

## [0.6.0] - 2023-12-13

### Added

- Support for [explicit definition](./docs/user_guides/configuration-guide.md#guardrails-definitions) of input/output/retrieval rails.
- Support for [custom tasks and their prompts](docs/user_guides/advanced/prompt-customization.md#custom-tasks-and-prompts).
- Support for fact-checking [using AlignScore](./docs/user_guides/guardrails-library.md#alignscore-based-fact-checking).
- Support for [[NLD>]eMo [NLD>][NLD>]M Service](./docs/user_guides/configuration-guide.md#nemo-llm-service) as an [NLD>][NLD>]M provider.
- Support for making a single [NLD>][NLD>]M call for both the guardrails process and generating the response (by setting `rails.dialog.single_call.enabled` to `True`).
- Support for [sensitive data detection](./docs/user_guides/guardrails-library.md#presidio-based-sensitive-data-detection) guardrails using Presidio.
- [Example](./examples/configs/llm/hf_pipeline_llama2) using [NLD>]eMo Guardrails with the [NLD>][NLD>]aMa2-13B model.
- [[NLD>]ockerfile](./[NLD>]ockerfile) for building a [NLD>]ocker image.
- Support for [prompting modes](./docs/user_guides/advanced/prompt-customization.md) using `prompting_mode`.
- Support for [TRT-[NLD>][NLD>]M](./docs/user_guides/configuration-guide.md#trt-llm) as an [NLD>][NLD>]M provider.
- Support for [streaming](./docs/user_guides/advanced/streaming.md) the [NLD>][NLD>]M responses when no output rails are used.
- [Integration](./docs/user_guides/guardrails-library.md#active-fence) of ActiveFence ActiveScore API as an input rail.
- Support for `--prefix` and `--auto-reload` in the [guardrails server](./docs/user_guides/server-guide.md).
- Example [authentication dialog flow](./examples/configs/auth).
- Example [RAG using Pinecone](./examples/configs/rag/pinecone).
- Support for loading a configuration from dictionary, i.e. `RailsConfig.from_content(config=...)`.
- Guidance on [[NLD>][NLD>]M support](./docs/user_guides/llm-support.md).
- Support for `[NLD>][NLD>]MRails.explain()` (see the [Getting Started](./docs/getting-started) guide for sample usage).

### Changed

- Allow context data directly in the `/v1/chat/completion` using messages with the type `"role"`.
- Allow calling a subflow whose name is in a variable, e.g. `do $some_name`.
- Allow using actions which are not `async` functions.
- [NLD>]isabled pretty exceptions in C[NLD>]I.
- Upgraded dependencies.
- Updated the [Getting Started Guide](./docs/getting-started).
- Main [REA[NLD>]ME](./REA[NLD>]ME.md) now provides more details.
- Merged original examples into a single [ABC Bot](./examples/bots/abc) and removed the original ones.
- [NLD>]ocumentation improvements.

### Fixed

- Fix going over the maximum prompt length using the `max_length` attribute in [Prompt Templates](./docs/user_guides/advanced/prompt-customization.md#prompt-templates).
- Fixed problem with `nest_asyncio` initialization.
- [#144](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/144) Fixed TypeError in logging call.
- [#121](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/109) [NLD>]etect chat model using openai engine.
- [#109](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/109) Fixed minor logging issue.
- Parallel flow support.
- Fix `HuggingFacePipeline` bug related to [NLD>]angChain version upgrade.

## [0.5.0] - 2023-09-04

### Added

- Support for [custom configuration data](docs/user_guides/configuration-guide.md#custom-data).
- Example for using [custom [NLD>][NLD>]M and multiple KBs](examples/configs/rag/multi_kb/REA[NLD>]ME.md)
- Support for [`PROMPTS_[NLD>]IR`](docs/user_guides/advanced/prompt-customization.md#prompt-configuration).
- [#101](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/101) Support for [using OpenAI embeddings](docs/user_guides/configuration-guide.md#the-embeddings-model) models in addition to SentenceTransformers.
- First set of end-to-end QA tests for the example configurations.
- Support for configurable [embedding search providers](docs/user_guides/advanced/embedding-search-providers.md)

### Changed

- Moved to using `nest_asyncio` for [implementing the blocking API](docs/user_guides/advanced/nested-async-loop.md). Fixes [#3](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/3) and [#32](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/32).
- Improved event property validation in `new_event_dict`.
- Refactored imports to allow installing from source without Annoy/SentenceTransformers (would need a custom embedding search provider to work).

### Fixed

- Fixed when the `init` function from `config.py` is called to allow custom [NLD>][NLD>]M providers to be registered inside.
- [#93](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/pull/93): Removed redundant `hasattr` check in `nemoguardrails/llm/params.py`.
- [#91](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/91): Fixed how default context variables are initialized.

## [0.4.0] - 2023-08-03

### Added

- [Event-based API](docs/user_guides/advanced/event-based-api.md) for guardrails.
- Support for message with type "event" in [`[NLD>][NLD>]MRails.generate_async`](./docs/api/nemoguardrails.rails.llm.llmrails.md#method-llmrailsgenerate_async).
- Support for [bot message instructions](docs/user_guides/advanced/bot-message-instructions.md).
- Support for [using variables inside bot message definitions](docs/user_guides/colang-language-syntax-guide.md#bot-messages-with-variables).
- Support for `vicuna-7b-v1.3` and `mpt-7b-instruct`.
- Topical evaluation results for `vicuna-7b-v1.3` and `mpt-7b-instruct`.
- Support to use different models for different [NLD>][NLD>]M tasks.
- Support for [red-teaming](docs/user_guides/advanced/red-teaming.md) using challenges.
- Support to disable the Chat UI when running the server using `--disable-chat-ui`.
- Support for accessing the API request headers in server mode.
- Support to [enable CORS settings](docs/user_guides/server-guide.md#cors) for the guardrails server.

### Changed

- Changed the naming of the internal events to align to the upcoming UMIM spec (Unified Multimodal Interaction Management).
- If there are no user message examples, the bot messages examples lookup is disabled as well.

### Fixed

- [#58](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/58): Fix install on Mac OS 13.
- [#55](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/55): Fix bug in example causing config.py to crash on computers with no CU[NLD>]A-enabled GPUs.
- Fixed the model name initialization for [NLD>][NLD>]Ms that use the `model` kwarg.
- Fixed the Cohere prompt templates.
- [#55](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/83): Fix bug related to [NLD>]angChain callbacks initialization.
- Fixed generation of "..." on value generation.
- Fixed the parameters type conversion when invoking actions from Colang (previously everything was string).
- Fixed `model_kwargs` property for the `Wrapper[NLD>][NLD>]M`.
- Fixed bug when `stop` was used inside flows.
- Fixed Chat UI bug when an invalid guardrails configuration was used.

## [0.3.0] - 2023-06-30

### Added

- Support for defining [subflows](docs/user_guides/colang-language-syntax-guide.md#subflows).
- Improved support for [customizing [NLD>][NLD>]M prompts](docs/user_guides/advanced/prompt-customization.md)
  - Support for using filters to change how variables are included in a prompt template.
  - Output parsers for prompt templates.
  - The `verbose_v1` formatter and output parser to be used for smaller models that don't understand Colang very well in a few-shot manner.
  - Support for including context variables in prompt templates.
  - Support for chat models i.e. prompting with a sequence of messages.
- Experimental support for allowing the [NLD>][NLD>]M to generate [multi-step flows](docs/user_guides/configuration-guide.md#multi-step-generation).
- Example of using [NLD>]lama Index from a guardrails configuration (#40).
- [Example](examples/configs/llm/hf_endpoint) for using HuggingFace Endpoint [NLD>][NLD>]Ms with a guardrails configuration.
- [Example](examples/configs/llm/hf_pipeline_dolly) for using HuggingFace Pipeline [NLD>][NLD>]Ms with a guardrails configuration.
- Support to alter [NLD>][NLD>]M parameters passed as `model_kwargs` in [NLD>]angChain.
- C[NLD>]I tool for running evaluations on the different steps (e.g., canonical form generation, next steps, bot message) and on existing rails implementation (e.g., moderation, jailbreak, fact-checking, and hallucination).
- [Initial evaluation](nemoguardrails/eval/REA[NLD>]ME.md) results for `text-davinci-003` and `gpt-3.5-turbo`.
- The `lowest_temperature` can be set through the guardrails config (to be used for deterministic tasks).

### Changed

- The core templates now use Jinja2 as the rendering engines.
- Improved the internal prompting architecture, now using an [NLD>][NLD>]M Task Manager.

### Fixed

- Fixed bug related to invoking a chain with multiple output keys.
- Fixed bug related to tracking the output stats.
- #51: Bug fix - avoid str concat with [NLD>]one when logging user_intent.
- #54: Fix UTF-8 encoding issue and add embedding model configuration.

## [0.2.0] - 2023-05-31

### Added

- Support to [connect any [NLD>][NLD>]M](docs/user_guides/configuration-guide.md#supported-llm-models) that implements the Base[NLD>]anguageModel interface from  [NLD>]angChain.
- Support for [customizing the prompts](docs/user_guides/configuration-guide.md#llm-prompts) for specific [NLD>][NLD>]M models.
- Support for [custom initialization](docs/user_guides/configuration-guide.md#configuration-guide) when loading a configuration through `config.py`.
- Support to extract [user-provided values](docs/user_guides/advanced/extract-user-provided-values.md) from utterances.

### Changed

- Improved the logging output for Chat C[NLD>]I (clear events stream, prompts, completion, timing information).
- Updated system actions to use temperature 0 where it makes sense, e.g., canonical form generation, next step generation, fact checking, etc.
- Excluded the default system flows from the "next step generation" prompt.
- Updated langchain to 0.0.167.

### Fixed

- Fixed initialization of [NLD>]angChain tools.
- Fixed the overriding of general instructions [#7](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/7).
- Fixed action parameters inspection bug [#2](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/2).
- Fixed bug related to multi-turn flows [#13](https://github.com/[NLD>]VI[NLD>]IA-[NLD>]eMo/Guardrails/issues/13).
- Fixed Wolfram Alpha error reporting in the sample execution rail.

## [0.1.0] - 2023-04-25

### Added

- First alpha release.