-
Notifications
You must be signed in to change notification settings - Fork 2.3k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Make format_instructions an optional argument in chat agent #2931
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This simple change allows users to fully overwrite the prompts when creating a chat agent, and optimize them for their use case. This minimal change makes the agent code much more generally useful. Previously, they could overwrite everything except for FORMAT_INSTRUCTIONS, which is essential to customization.
The latest updates on your projects. Learn more about Vercel for Git ↗︎
|
dosubot
bot
added
the
auto:improvement
Medium size change to existing code to handle new use-cases
label
Oct 16, 2023
Seems good to me, thanks! |
nicolewhite
referenced
this pull request
in autoblocksai/javascript-sdk
Oct 31, 2023
[![Mend Renovate](https://app.renovatebot.com/images/banner.svg)](https://renovatebot.com) This PR contains the following updates: | Package | Change | Age | Adoption | Passing | Confidence | |---|---|---|---|---|---| | [langchain](https://togithub.com/langchain-ai/langchainjs) | [`^0.0.165` -> `^0.0.175`](https://renovatebot.com/diffs/npm/langchain/0.0.165/0.0.175) | [![age](https://developer.mend.io/api/mc/badges/age/npm/langchain/0.0.175?slim=true)](https://docs.renovatebot.com/merge-confidence/) | [![adoption](https://developer.mend.io/api/mc/badges/adoption/npm/langchain/0.0.175?slim=true)](https://docs.renovatebot.com/merge-confidence/) | [![passing](https://developer.mend.io/api/mc/badges/compatibility/npm/langchain/0.0.165/0.0.175?slim=true)](https://docs.renovatebot.com/merge-confidence/) | [![confidence](https://developer.mend.io/api/mc/badges/confidence/npm/langchain/0.0.165/0.0.175?slim=true)](https://docs.renovatebot.com/merge-confidence/) | --- ### Release Notes <details> <summary>langchain-ai/langchainjs (langchain)</summary> ### [`v0.0.175`](https://togithub.com/langchain-ai/langchainjs/releases/tag/0.0.175) [Compare Source](https://togithub.com/langchain-ai/langchainjs/compare/0.0.174...0.0.175) #### What's Changed - Release 0.0.174 by [@​jacoblee93](https://togithub.com/jacoblee93) in [https://github.com/langchain-ai/langchainjs/pull/3064](https://togithub.com/langchain-ai/langchainjs/pull/3064) - Adds basic critique revise cookbook by [@​jacoblee93](https://togithub.com/jacoblee93) in [https://github.com/langchain-ai/langchainjs/pull/3067](https://togithub.com/langchain-ai/langchainjs/pull/3067) - fix: calculate Milvus textlength in bytes by [@​kurtmoser](https://togithub.com/kurtmoser) in [https://github.com/langchain-ai/langchainjs/pull/3079](https://togithub.com/langchain-ai/langchainjs/pull/3079) - Add Convex integration by [@​xixixao](https://togithub.com/xixixao) in [https://github.com/langchain-ai/langchainjs/pull/2977](https://togithub.com/langchain-ai/langchainjs/pull/2977) - Make "IN-filters" possible for Vercel Postgres Metadata by [@​rbouschery](https://togithub.com/rbouschery) in [https://github.com/langchain-ai/langchainjs/pull/3040](https://togithub.com/langchain-ai/langchainjs/pull/3040) - Fix Convex typo by [@​jacoblee93](https://togithub.com/jacoblee93) in [https://github.com/langchain-ai/langchainjs/pull/3081](https://togithub.com/langchain-ai/langchainjs/pull/3081) #### New Contributors - [@​kurtmoser](https://togithub.com/kurtmoser) made their first contribution in [https://github.com/langchain-ai/langchainjs/pull/3079](https://togithub.com/langchain-ai/langchainjs/pull/3079) - [@​xixixao](https://togithub.com/xixixao) made their first contribution in [https://github.com/langchain-ai/langchainjs/pull/2977](https://togithub.com/langchain-ai/langchainjs/pull/2977) - [@​rbouschery](https://togithub.com/rbouschery) made their first contribution in [https://github.com/langchain-ai/langchainjs/pull/3040](https://togithub.com/langchain-ai/langchainjs/pull/3040) **Full Changelog**: langchain-ai/langchainjs@0.0.174...0.0.175 ### [`v0.0.174`](https://togithub.com/langchain-ai/langchainjs/releases/tag/0.0.174) [Compare Source](https://togithub.com/langchain-ai/langchainjs/compare/0.0.173...0.0.174) #### What's Changed - Release 0.0.173 by [@​jacoblee93](https://togithub.com/jacoblee93) in [https://github.com/langchain-ai/langchainjs/pull/3054](https://togithub.com/langchain-ai/langchainjs/pull/3054) - fix: LANGCHAIN_VERBOSE=false does not turn off verbose logging. by [@​xianminx](https://togithub.com/xianminx) in [https://github.com/langchain-ai/langchainjs/pull/3057](https://togithub.com/langchain-ai/langchainjs/pull/3057) - Format by [@​jacoblee93](https://togithub.com/jacoblee93) in [https://github.com/langchain-ai/langchainjs/pull/3058](https://togithub.com/langchain-ai/langchainjs/pull/3058) - fix: improper markdown links by [@​bracesproul](https://togithub.com/bracesproul) in [https://github.com/langchain-ai/langchainjs/pull/3061](https://togithub.com/langchain-ai/langchainjs/pull/3061) - Adds MMR search for Pinecone by [@​jacoblee93](https://togithub.com/jacoblee93) in [https://github.com/langchain-ai/langchainjs/pull/3059](https://togithub.com/langchain-ai/langchainjs/pull/3059) - Allow vectorstore retrievers in SemanticSimilarityExampleSelector.ts by [@​jacoblee93](https://togithub.com/jacoblee93) in [https://github.com/langchain-ai/langchainjs/pull/3056](https://togithub.com/langchain-ai/langchainjs/pull/3056) - Add getOutputValue to langchain/memory exports by [@​mattmcgiv](https://togithub.com/mattmcgiv) in [https://github.com/langchain-ai/langchainjs/pull/3060](https://togithub.com/langchain-ai/langchainjs/pull/3060) #### New Contributors - [@​xianminx](https://togithub.com/xianminx) made their first contribution in [https://github.com/langchain-ai/langchainjs/pull/3057](https://togithub.com/langchain-ai/langchainjs/pull/3057) - [@​mattmcgiv](https://togithub.com/mattmcgiv) made their first contribution in [https://github.com/langchain-ai/langchainjs/pull/3060](https://togithub.com/langchain-ai/langchainjs/pull/3060) **Full Changelog**: langchain-ai/langchainjs@0.0.173...0.0.174 ### [`v0.0.173`](https://togithub.com/langchain-ai/langchainjs/releases/tag/0.0.173) [Compare Source](https://togithub.com/langchain-ai/langchainjs/compare/0.0.172...0.0.173) #### What's Changed - Release 0.0.172 by [@​jacoblee93](https://togithub.com/jacoblee93) in [https://github.com/langchain-ai/langchainjs/pull/3030](https://togithub.com/langchain-ai/langchainjs/pull/3030) - feat: base storage docs by [@​bracesproul](https://togithub.com/bracesproul) in [https://github.com/langchain-ai/langchainjs/pull/2984](https://togithub.com/langchain-ai/langchainjs/pull/2984) - Refactor docs, add Scrimba course by [@​jacoblee93](https://togithub.com/jacoblee93) in [https://github.com/langchain-ai/langchainjs/pull/3032](https://togithub.com/langchain-ai/langchainjs/pull/3032) - Hide storage docs for now by [@​jacoblee93](https://togithub.com/jacoblee93) in [https://github.com/langchain-ai/langchainjs/pull/3034](https://togithub.com/langchain-ai/langchainjs/pull/3034) - Update broken link on SerpApi documentation by [@​hilmanski](https://togithub.com/hilmanski) in [https://github.com/langchain-ai/langchainjs/pull/3033](https://togithub.com/langchain-ai/langchainjs/pull/3033) - Fix broken docs links by [@​jacoblee93](https://togithub.com/jacoblee93) in [https://github.com/langchain-ai/langchainjs/pull/3035](https://togithub.com/langchain-ai/langchainjs/pull/3035) - feat: added openai functions output parser class by [@​bracesproul](https://togithub.com/bracesproul) in [https://github.com/langchain-ai/langchainjs/pull/3029](https://togithub.com/langchain-ai/langchainjs/pull/3029) - feat: added agent xml output parser class by [@​bracesproul](https://togithub.com/bracesproul) in [https://github.com/langchain-ai/langchainjs/pull/3037](https://togithub.com/langchain-ai/langchainjs/pull/3037) - Use node: prefixed deps by [@​jacoblee93](https://togithub.com/jacoblee93) in [https://github.com/langchain-ai/langchainjs/pull/3038](https://togithub.com/langchain-ai/langchainjs/pull/3038) - Add cookbook section and rewrite-retrieve-read example by [@​jacoblee93](https://togithub.com/jacoblee93) in [https://github.com/langchain-ai/langchainjs/pull/3039](https://togithub.com/langchain-ai/langchainjs/pull/3039) - Update deps in cookbook by [@​jacoblee93](https://togithub.com/jacoblee93) in [https://github.com/langchain-ai/langchainjs/pull/3046](https://togithub.com/langchain-ai/langchainjs/pull/3046) - fix: remove .js from imports in examples by [@​bracesproul](https://togithub.com/bracesproul) in [https://github.com/langchain-ai/langchainjs/pull/3047](https://togithub.com/langchain-ai/langchainjs/pull/3047) - Fix: bug where pgvector cant add more than chunk size by [@​bracesproul](https://togithub.com/bracesproul) in [https://github.com/langchain-ai/langchainjs/pull/3049](https://togithub.com/langchain-ai/langchainjs/pull/3049) - Add new scrims to docs by [@​jacoblee93](https://togithub.com/jacoblee93) in [https://github.com/langchain-ai/langchainjs/pull/3050](https://togithub.com/langchain-ai/langchainjs/pull/3050) - Use prepared statements + add concurrency for Cassandra by [@​mieslep](https://togithub.com/mieslep) in [https://github.com/langchain-ai/langchainjs/pull/3043](https://togithub.com/langchain-ai/langchainjs/pull/3043) - fix: add util for formatting docs page content by [@​bracesproul](https://togithub.com/bracesproul) in [https://github.com/langchain-ai/langchainjs/pull/3048](https://togithub.com/langchain-ai/langchainjs/pull/3048) - Feat: runnable examples in agent how to docs by [@​bracesproul](https://togithub.com/bracesproul) in [https://github.com/langchain-ai/langchainjs/pull/3027](https://togithub.com/langchain-ai/langchainjs/pull/3027) - feat: rag fusion notebook by [@​bracesproul](https://togithub.com/bracesproul) in [https://github.com/langchain-ai/langchainjs/pull/3051](https://togithub.com/langchain-ai/langchainjs/pull/3051) #### New Contributors - [@​hilmanski](https://togithub.com/hilmanski) made their first contribution in [https://github.com/langchain-ai/langchainjs/pull/3033](https://togithub.com/langchain-ai/langchainjs/pull/3033) - [@​mieslep](https://togithub.com/mieslep) made their first contribution in [https://github.com/langchain-ai/langchainjs/pull/3043](https://togithub.com/langchain-ai/langchainjs/pull/3043) **Full Changelog**: langchain-ai/langchainjs@0.0.172...0.0.173 ### [`v0.0.172`](https://togithub.com/langchain-ai/langchainjs/releases/tag/0.0.172) [Compare Source](https://togithub.com/langchain-ai/langchainjs/compare/0.0.171...0.0.172) #### What's Changed - Release 0.0.171 by [@​jacoblee93](https://togithub.com/jacoblee93) in [https://github.com/langchain-ai/langchainjs/pull/3019](https://togithub.com/langchain-ai/langchainjs/pull/3019) - Avoid hardcoding embedding size to OpenAI dimensions in Qdrant client by [@​jacoblee93](https://togithub.com/jacoblee93) in [https://github.com/langchain-ai/langchainjs/pull/3024](https://togithub.com/langchain-ai/langchainjs/pull/3024) - Remove more hard-coded values by [@​jacoblee93](https://togithub.com/jacoblee93) in [https://github.com/langchain-ai/langchainjs/pull/3025](https://togithub.com/langchain-ai/langchainjs/pull/3025) - feat: added react output parser and tests by [@​bracesproul](https://togithub.com/bracesproul) in [https://github.com/langchain-ai/langchainjs/pull/3020](https://togithub.com/langchain-ai/langchainjs/pull/3020) - Feat: generic file system kv store by [@​bracesproul](https://togithub.com/bracesproul) in [https://github.com/langchain-ai/langchainjs/pull/3013](https://togithub.com/langchain-ai/langchainjs/pull/3013) - Respect organization in OpenAIEmbeddings by [@​jacoblee93](https://togithub.com/jacoblee93) in [https://github.com/langchain-ai/langchainjs/pull/3028](https://togithub.com/langchain-ai/langchainjs/pull/3028) **Full Changelog**: langchain-ai/langchainjs@0.0.171...0.0.172 ### [`v0.0.171`](https://togithub.com/langchain-ai/langchainjs/releases/tag/0.0.171) [Compare Source](https://togithub.com/langchain-ai/langchainjs/compare/0.0.170...0.0.171) #### What's Changed - Release 0.0.170 by [@​jacoblee93](https://togithub.com/jacoblee93) in [https://github.com/langchain-ai/langchainjs/pull/2996](https://togithub.com/langchain-ai/langchainjs/pull/2996) - fix: updated title of docs and file name by [@​bracesproul](https://togithub.com/bracesproul) in [https://github.com/langchain-ai/langchainjs/pull/2997](https://togithub.com/langchain-ai/langchainjs/pull/2997) - Chore/neo4j security note by [@​bracesproul](https://togithub.com/bracesproul) in [https://github.com/langchain-ai/langchainjs/pull/2998](https://togithub.com/langchain-ai/langchainjs/pull/2998) - IterableReadableStream.fromReadableStream reader releaseLock by [@​AIXGEEK](https://togithub.com/AIXGEEK) in [https://github.com/langchain-ai/langchainjs/pull/3003](https://togithub.com/langchain-ai/langchainjs/pull/3003) - Ensure shell expansions are quoted in docs workflows. by [@​obi1kenobi](https://togithub.com/obi1kenobi) in [https://github.com/langchain-ai/langchainjs/pull/3014](https://togithub.com/langchain-ai/langchainjs/pull/3014) - Upgrade `actions/checkout` to v4 in all workflows. by [@​obi1kenobi](https://togithub.com/obi1kenobi) in [https://github.com/langchain-ai/langchainjs/pull/3015](https://togithub.com/langchain-ai/langchainjs/pull/3015) - fix top_p value by [@​hemanth](https://togithub.com/hemanth) in [https://github.com/langchain-ai/langchainjs/pull/3008](https://togithub.com/langchain-ai/langchainjs/pull/3008) - nit: add missing eslint disable any to runnable passthrough by [@​bracesproul](https://togithub.com/bracesproul) in [https://github.com/langchain-ai/langchainjs/pull/3017](https://togithub.com/langchain-ai/langchainjs/pull/3017) - Feat: added runnable examples to agent docs by [@​bracesproul](https://togithub.com/bracesproul) in [https://github.com/langchain-ai/langchainjs/pull/3000](https://togithub.com/langchain-ai/langchainjs/pull/3000) - WebPDFLoader: allow usage of custom pdfjs build by [@​DennisKo](https://togithub.com/DennisKo) in [https://github.com/langchain-ai/langchainjs/pull/3010](https://togithub.com/langchain-ai/langchainjs/pull/3010) - Feat/parent doc score threshold by [@​bracesproul](https://togithub.com/bracesproul) in [https://github.com/langchain-ai/langchainjs/pull/3011](https://togithub.com/langchain-ai/langchainjs/pull/3011) - Llama2 node-llama-cpp chat module by [@​nigel-daniels](https://togithub.com/nigel-daniels) in [https://github.com/langchain-ai/langchainjs/pull/2999](https://togithub.com/langchain-ai/langchainjs/pull/2999) - Add support for cassandra filtering, update docs by [@​alexleventer](https://togithub.com/alexleventer) in [https://github.com/langchain-ai/langchainjs/pull/2967](https://togithub.com/langchain-ai/langchainjs/pull/2967) - feat: update mvi integration for more metadata types by [@​malandis](https://togithub.com/malandis) in [https://github.com/langchain-ai/langchainjs/pull/3002](https://togithub.com/langchain-ai/langchainjs/pull/3002) #### New Contributors - [@​AIXGEEK](https://togithub.com/AIXGEEK) made their first contribution in [https://github.com/langchain-ai/langchainjs/pull/3003](https://togithub.com/langchain-ai/langchainjs/pull/3003) - [@​hemanth](https://togithub.com/hemanth) made their first contribution in [https://github.com/langchain-ai/langchainjs/pull/3008](https://togithub.com/langchain-ai/langchainjs/pull/3008) - [@​DennisKo](https://togithub.com/DennisKo) made their first contribution in [https://github.com/langchain-ai/langchainjs/pull/3010](https://togithub.com/langchain-ai/langchainjs/pull/3010) **Full Changelog**: langchain-ai/langchainjs@0.0.170...0.0.171 ### [`v0.0.170`](https://togithub.com/langchain-ai/langchainjs/releases/tag/0.0.170) [Compare Source](https://togithub.com/langchain-ai/langchainjs/compare/0.0.169...0.0.170) ##### What's Changed - Release 0.0.169 by [@​jacoblee93](https://togithub.com/jacoblee93) in [https://github.com/langchain-ai/langchainjs/pull/2971](https://togithub.com/langchain-ai/langchainjs/pull/2971) - fix: show voe notebook in docs by [@​bracesproul](https://togithub.com/bracesproul) in [https://github.com/langchain-ai/langchainjs/pull/2973](https://togithub.com/langchain-ai/langchainjs/pull/2973) - Revert "fix: show voe notebook in docs" by [@​bracesproul](https://togithub.com/bracesproul) in [https://github.com/langchain-ai/langchainjs/pull/2974](https://togithub.com/langchain-ai/langchainjs/pull/2974) - fix: wildcard suffix for yielding keys by [@​bracesproul](https://togithub.com/bracesproul) in [https://github.com/langchain-ai/langchainjs/pull/2980](https://togithub.com/langchain-ai/langchainjs/pull/2980) - Feat/security notice openapi by [@​bracesproul](https://togithub.com/bracesproul) in [https://github.com/langchain-ai/langchainjs/pull/2959](https://togithub.com/langchain-ai/langchainjs/pull/2959) - feat: SQL chain security disclaimers to documentation & JSDoc by [@​bracesproul](https://togithub.com/bracesproul) in [https://github.com/langchain-ai/langchainjs/pull/2962](https://togithub.com/langchain-ai/langchainjs/pull/2962) - fix: update links to langchain org by [@​bracesproul](https://togithub.com/bracesproul) in [https://github.com/langchain-ai/langchainjs/pull/2983](https://togithub.com/langchain-ai/langchainjs/pull/2983) - Fixed order of LCEL menu and added tables in interface page by [@​tsensei](https://togithub.com/tsensei) in [https://github.com/langchain-ai/langchainjs/pull/2986](https://togithub.com/langchain-ai/langchainjs/pull/2986) - Add exports, small CI fix by [@​jacoblee93](https://togithub.com/jacoblee93) in [https://github.com/langchain-ai/langchainjs/pull/2993](https://togithub.com/langchain-ai/langchainjs/pull/2993) - Sveltekit - deployment guide by [@​SimonPrammer](https://togithub.com/SimonPrammer) in [https://github.com/langchain-ai/langchainjs/pull/2987](https://togithub.com/langchain-ai/langchainjs/pull/2987) - feat: added invoke method to chat model docs by [@​bracesproul](https://togithub.com/bracesproul) in [https://github.com/langchain-ai/langchainjs/pull/2981](https://togithub.com/langchain-ai/langchainjs/pull/2981) - fix: add missing codeblock import by [@​bracesproul](https://togithub.com/bracesproul) in [https://github.com/langchain-ai/langchainjs/pull/2994](https://togithub.com/langchain-ai/langchainjs/pull/2994) - Run format on docs PRs by [@​jacoblee93](https://togithub.com/jacoblee93) in [https://github.com/langchain-ai/langchainjs/pull/2995](https://togithub.com/langchain-ai/langchainjs/pull/2995) - Add ERNIE-Bot-4 from Baidu Wenxin Qianfan by [@​sunner](https://togithub.com/sunner) in [https://github.com/langchain-ai/langchainjs/pull/2969](https://togithub.com/langchain-ai/langchainjs/pull/2969) - Feat/upstash redis storage by [@​bracesproul](https://togithub.com/bracesproul) in [https://github.com/langchain-ai/langchainjs/pull/2963](https://togithub.com/langchain-ai/langchainjs/pull/2963) - feat: runnables powered agent by [@​bracesproul](https://togithub.com/bracesproul) in [https://github.com/langchain-ai/langchainjs/pull/2927](https://togithub.com/langchain-ai/langchainjs/pull/2927) - Feat: runnable assign by [@​bracesproul](https://togithub.com/bracesproul) in [https://github.com/langchain-ai/langchainjs/pull/2948](https://togithub.com/langchain-ai/langchainjs/pull/2948) - fix: ensure metadata is cast to jsonb for pgvector by [@​devinburnette](https://togithub.com/devinburnette) in [https://github.com/langchain-ai/langchainjs/pull/2949](https://togithub.com/langchain-ai/langchainjs/pull/2949) - Get transcript text by ID in AssemblyAI by [@​Swimburger](https://togithub.com/Swimburger) in [https://github.com/langchain-ai/langchainjs/pull/2582](https://togithub.com/langchain-ai/langchainjs/pull/2582) ##### New Contributors - [@​devinburnette](https://togithub.com/devinburnette) made their first contribution in [https://github.com/langchain-ai/langchainjs/pull/2949](https://togithub.com/langchain-ai/langchainjs/pull/2949) **Full Changelog**: langchain-ai/langchainjs@0.0.169...0.0.170 ### [`v0.0.169`](https://togithub.com/langchain-ai/langchainjs/releases/tag/0.0.169) [Compare Source](https://togithub.com/langchain-ai/langchainjs/compare/0.0.168...0.0.169) ##### What's Changed - Release 0.0.168 by [@​jacoblee93](https://togithub.com/jacoblee93) in [https://github.com/langchain-ai/langchainjs/pull/2950](https://togithub.com/langchain-ai/langchainjs/pull/2950) - Add build + docs deploy step in GHA by [@​jacoblee93](https://togithub.com/jacoblee93) in [https://github.com/langchain-ai/langchainjs/pull/2957](https://togithub.com/langchain-ai/langchainjs/pull/2957) - Fixed typo in documentation by [@​tsensei](https://togithub.com/tsensei) in [https://github.com/langchain-ai/langchainjs/pull/2952](https://togithub.com/langchain-ai/langchainjs/pull/2952) - Feat: general security doc by [@​bracesproul](https://togithub.com/bracesproul) in [https://github.com/langchain-ai/langchainjs/pull/2958](https://togithub.com/langchain-ai/langchainjs/pull/2958) - zep: Add native MMR to Zep Retriever by [@​danielchalef](https://togithub.com/danielchalef) in [https://github.com/langchain-ai/langchainjs/pull/2955](https://togithub.com/langchain-ai/langchainjs/pull/2955) - add delete for faiss by [@​ewfian](https://togithub.com/ewfian) in [https://github.com/langchain-ai/langchainjs/pull/2924](https://togithub.com/langchain-ai/langchainjs/pull/2924) - Weaviate - throw error if storing vectors fails by [@​the-powerpointer](https://togithub.com/the-powerpointer) in [https://github.com/langchain-ai/langchainjs/pull/2942](https://togithub.com/langchain-ai/langchainjs/pull/2942) - Add missing CI action env by [@​jacoblee93](https://togithub.com/jacoblee93) in [https://github.com/langchain-ai/langchainjs/pull/2965](https://togithub.com/langchain-ai/langchainjs/pull/2965) - LLMonitor Callback Handler: fix tracking users on agents by [@​vincelwt](https://togithub.com/vincelwt) in [https://github.com/langchain-ai/langchainjs/pull/2941](https://togithub.com/langchain-ai/langchainjs/pull/2941) - Add YandexGPT Chat model by [@​tyumentsev4](https://togithub.com/tyumentsev4) in [https://github.com/langchain-ai/langchainjs/pull/2930](https://togithub.com/langchain-ai/langchainjs/pull/2930) - Remove preview deploy for now by [@​jacoblee93](https://togithub.com/jacoblee93) in [https://github.com/langchain-ai/langchainjs/pull/2968](https://togithub.com/langchain-ai/langchainjs/pull/2968) - Feat: violations of expectations chain by [@​bracesproul](https://togithub.com/bracesproul) in [https://github.com/langchain-ai/langchainjs/pull/2945](https://togithub.com/langchain-ai/langchainjs/pull/2945) - Adds base URL support for Cloudflare WorkersAI by [@​jacoblee93](https://togithub.com/jacoblee93) in [https://github.com/langchain-ai/langchainjs/pull/2970](https://togithub.com/langchain-ai/langchainjs/pull/2970) ##### New Contributors - [@​tsensei](https://togithub.com/tsensei) made their first contribution in [https://github.com/langchain-ai/langchainjs/pull/2952](https://togithub.com/langchain-ai/langchainjs/pull/2952) **Full Changelog**: langchain-ai/langchainjs@0.0.168...0.0.169 ### [`v0.0.168`](https://togithub.com/langchain-ai/langchainjs/releases/tag/0.0.168) [Compare Source](https://togithub.com/langchain-ai/langchainjs/compare/0.0.167...0.0.168) #### What's Changed - Release 0.0.167 by [@​jacoblee93](https://togithub.com/jacoblee93) in [https://github.com/langchain-ai/langchainjs/pull/2916](https://togithub.com/langchain-ai/langchainjs/pull/2916) - Mark the docs scripts files as executable. by [@​obi1kenobi](https://togithub.com/obi1kenobi) in [https://github.com/langchain-ai/langchainjs/pull/2932](https://togithub.com/langchain-ai/langchainjs/pull/2932) - Add more memory to linter by [@​jacoblee93](https://togithub.com/jacoblee93) in [https://github.com/langchain-ai/langchainjs/pull/2934](https://togithub.com/langchain-ai/langchainjs/pull/2934) - Update repo path to point to `langchain-ai` GitHub org. by [@​obi1kenobi](https://togithub.com/obi1kenobi) in [https://github.com/langchain-ai/langchainjs/pull/2933](https://togithub.com/langchain-ai/langchainjs/pull/2933) - fix: update llm_streaming_stream_method.ts doc by [@​cjbt](https://togithub.com/cjbt) in [https://github.com/langchain-ai/langchainjs/pull/2929](https://togithub.com/langchain-ai/langchainjs/pull/2929) - Make format_instructions an optional argument in chat agent by [@​EKMeyerson](https://togithub.com/EKMeyerson) in [https://github.com/langchain-ai/langchainjs/pull/2931](https://togithub.com/langchain-ai/langchainjs/pull/2931) - Update graph_db_custom_prompt.ts by [@​tomasonjo](https://togithub.com/tomasonjo) in [https://github.com/langchain-ai/langchainjs/pull/2920](https://togithub.com/langchain-ai/langchainjs/pull/2920) - Simplify docs structure, remove skeleton by [@​jacoblee93](https://togithub.com/jacoblee93) in [https://github.com/langchain-ai/langchainjs/pull/2935](https://togithub.com/langchain-ai/langchainjs/pull/2935) - Run build in series by [@​jacoblee93](https://togithub.com/jacoblee93) in [https://github.com/langchain-ai/langchainjs/pull/2936](https://togithub.com/langchain-ai/langchainjs/pull/2936) - Remove snippets by [@​jacoblee93](https://togithub.com/jacoblee93) in [https://github.com/langchain-ai/langchainjs/pull/2937](https://togithub.com/langchain-ai/langchainjs/pull/2937) - Use SWC to generate AST by [@​jacoblee93](https://togithub.com/jacoblee93) in [https://github.com/langchain-ai/langchainjs/pull/2939](https://togithub.com/langchain-ai/langchainjs/pull/2939) - WIP: add runManager.handleLLMNewToken missing call by [@​ivoneijr](https://togithub.com/ivoneijr) in [https://github.com/langchain-ai/langchainjs/pull/2910](https://togithub.com/langchain-ai/langchainjs/pull/2910) - Run exports tests in parallel on GitHub CI. by [@​obi1kenobi](https://togithub.com/obi1kenobi) in [https://github.com/langchain-ai/langchainjs/pull/2944](https://togithub.com/langchain-ai/langchainjs/pull/2944) - feat: add CloseVector as VectorStore provider by [@​jo32](https://togithub.com/jo32) in [https://github.com/langchain-ai/langchainjs/pull/2832](https://togithub.com/langchain-ai/langchainjs/pull/2832) - Feature: Add Fake integration support to match Python library by [@​frankolson](https://togithub.com/frankolson) in [https://github.com/langchain-ai/langchainjs/pull/2928](https://togithub.com/langchain-ai/langchainjs/pull/2928) - Adds Cloudflare Workers AI LLM and chat model by [@​jacoblee93](https://togithub.com/jacoblee93) in [https://github.com/langchain-ai/langchainjs/pull/2923](https://togithub.com/langchain-ai/langchainjs/pull/2923) #### New Contributors - [@​cjbt](https://togithub.com/cjbt) made their first contribution in [https://github.com/langchain-ai/langchainjs/pull/2929](https://togithub.com/langchain-ai/langchainjs/pull/2929) - [@​EKMeyerson](https://togithub.com/EKMeyerson) made their first contribution in [https://github.com/langchain-ai/langchainjs/pull/2931](https://togithub.com/langchain-ai/langchainjs/pull/2931) - [@​jo32](https://togithub.com/jo32) made their first contribution in [https://github.com/langchain-ai/langchainjs/pull/2832](https://togithub.com/langchain-ai/langchainjs/pull/2832) - [@​frankolson](https://togithub.com/frankolson) made their first contribution in [https://github.com/langchain-ai/langchainjs/pull/2928](https://togithub.com/langchain-ai/langchainjs/pull/2928) **Full Changelog**: langchain-ai/langchainjs@0.0.167...0.0.168 ### [`v0.0.167`](https://togithub.com/langchain-ai/langchainjs/releases/tag/0.0.167) [Compare Source](https://togithub.com/langchain-ai/langchainjs/compare/0.0.166...0.0.167) #### What's Changed - Release 0.0.166 by [@​jacoblee93](https://togithub.com/jacoblee93) in [https://github.com/langchain-ai/langchainjs/pull/2894](https://togithub.com/langchain-ai/langchainjs/pull/2894) - elasticsearch: customize user-agent for es client by [@​TattdCodeMonkey](https://togithub.com/TattdCodeMonkey) in [https://github.com/langchain-ai/langchainjs/pull/2897](https://togithub.com/langchain-ai/langchainjs/pull/2897) - Chore: update SQL popular chain docs to use runnables by [@​bracesproul](https://togithub.com/bracesproul) in [https://github.com/langchain-ai/langchainjs/pull/2877](https://togithub.com/langchain-ai/langchainjs/pull/2877) - fix: hide from sidebar by [@​bracesproul](https://togithub.com/bracesproul) in [https://github.com/langchain-ai/langchainjs/pull/2909](https://togithub.com/langchain-ai/langchainjs/pull/2909) - Chore: conversational QA docs to use Runnables by [@​bracesproul](https://togithub.com/bracesproul) in [https://github.com/langchain-ai/langchainjs/pull/2890](https://togithub.com/langchain-ai/langchainjs/pull/2890) - refactor: openai function calling docs by [@​bracesproul](https://togithub.com/bracesproul) in [https://github.com/langchain-ai/langchainjs/pull/2893](https://togithub.com/langchain-ai/langchainjs/pull/2893) - chore: added why use LCEL doc by [@​bracesproul](https://togithub.com/bracesproul) in [https://github.com/langchain-ai/langchainjs/pull/2900](https://togithub.com/langchain-ai/langchainjs/pull/2900) - fix: batch embedding queries, fix get client in tests by [@​bracesproul](https://togithub.com/bracesproul) in [https://github.com/langchain-ai/langchainjs/pull/2908](https://togithub.com/langchain-ai/langchainjs/pull/2908) - Improve Cassandra vector store docs by [@​alexleventer](https://togithub.com/alexleventer) in [https://github.com/langchain-ai/langchainjs/pull/2903](https://togithub.com/langchain-ai/langchainjs/pull/2903) - feat: Add Neo4j vector store by [@​easwee](https://togithub.com/easwee) in [https://github.com/langchain-ai/langchainjs/pull/2852](https://togithub.com/langchain-ai/langchainjs/pull/2852) - Add YandexGPT LLM model by [@​tyumentsev4](https://togithub.com/tyumentsev4) in [https://github.com/langchain-ai/langchainjs/pull/2906](https://togithub.com/langchain-ai/langchainjs/pull/2906) - Allow ParentDocumentRetriever to subclass MultiVectorRetriever by [@​jacoblee93](https://togithub.com/jacoblee93) in [https://github.com/langchain-ai/langchainjs/pull/2853](https://togithub.com/langchain-ai/langchainjs/pull/2853) - Add neo4j vector index docs by [@​tomasonjo](https://togithub.com/tomasonjo) in [https://github.com/langchain-ai/langchainjs/pull/2895](https://togithub.com/langchain-ai/langchainjs/pull/2895) - Docs updates by [@​jacoblee93](https://togithub.com/jacoblee93) in [https://github.com/langchain-ai/langchainjs/pull/2915](https://togithub.com/langchain-ai/langchainjs/pull/2915) - Add Vercel KV base store by [@​jacoblee93](https://togithub.com/jacoblee93) in [https://github.com/langchain-ai/langchainjs/pull/2912](https://togithub.com/langchain-ai/langchainjs/pull/2912) #### New Contributors - [@​TattdCodeMonkey](https://togithub.com/TattdCodeMonkey) made their first contribution in [https://github.com/langchain-ai/langchainjs/pull/2897](https://togithub.com/langchain-ai/langchainjs/pull/2897) **Full Changelog**: langchain-ai/langchainjs@0.0.166...0.0.167 ### [`v0.0.166`](https://togithub.com/langchain-ai/langchainjs/releases/tag/0.0.166) [Compare Source](https://togithub.com/langchain-ai/langchainjs/compare/0.0.165...0.0.166) #### What's Changed - Release 0.0.165 by [@​jacoblee93](https://togithub.com/jacoblee93) in [https://github.com/langchain-ai/langchainjs/pull/2870](https://togithub.com/langchain-ai/langchainjs/pull/2870) - Update docs for model streaming by [@​jacoblee93](https://togithub.com/jacoblee93) in [https://github.com/langchain-ai/langchainjs/pull/2887](https://togithub.com/langchain-ai/langchainjs/pull/2887) - Remote Runnables by [@​efriis](https://togithub.com/efriis) in [https://github.com/langchain-ai/langchainjs/pull/2828](https://togithub.com/langchain-ai/langchainjs/pull/2828) - added force default filter option to mergeFilters by [@​valdo99](https://togithub.com/valdo99) in [https://github.com/langchain-ai/langchainjs/pull/2882](https://togithub.com/langchain-ai/langchainjs/pull/2882) - Implement max marginal relevance search for Supabase vector store by [@​carlobeltrame](https://togithub.com/carlobeltrame) in [https://github.com/langchain-ai/langchainjs/pull/2800](https://togithub.com/langchain-ai/langchainjs/pull/2800) - feat: add momento vector index as a vector store provider by [@​malandis](https://togithub.com/malandis) in [https://github.com/langchain-ai/langchainjs/pull/2851](https://togithub.com/langchain-ai/langchainjs/pull/2851) - Fix: validate input or throw when using MessagesPlaceholder by [@​bracesproul](https://togithub.com/bracesproul) in [https://github.com/langchain-ai/langchainjs/pull/2874](https://togithub.com/langchain-ai/langchainjs/pull/2874) - Feat: added new advanced convo qa runnables use case doc by [@​bracesproul](https://togithub.com/bracesproul) in [https://github.com/langchain-ai/langchainjs/pull/2889](https://togithub.com/langchain-ai/langchainjs/pull/2889) - Unstructured API - additional props by [@​perstarkse](https://togithub.com/perstarkse) in [https://github.com/langchain-ai/langchainjs/pull/2888](https://togithub.com/langchain-ai/langchainjs/pull/2888) - Add AnthropicFunctionsBedrock by [@​efriis](https://togithub.com/efriis) in [https://github.com/langchain-ai/langchainjs/pull/2892](https://togithub.com/langchain-ai/langchainjs/pull/2892) #### New Contributors - [@​valdo99](https://togithub.com/valdo99) made their first contribution in [https://github.com/langchain-ai/langchainjs/pull/2882](https://togithub.com/langchain-ai/langchainjs/pull/2882) - [@​carlobeltrame](https://togithub.com/carlobeltrame) made their first contribution in [https://github.com/langchain-ai/langchainjs/pull/2800](https://togithub.com/langchain-ai/langchainjs/pull/2800) - [@​perstarkse](https://togithub.com/perstarkse) made their first contribution in [https://github.com/langchain-ai/langchainjs/pull/2888](https://togithub.com/langchain-ai/langchainjs/pull/2888) **Full Changelog**: langchain-ai/langchainjs@0.0.165...0.0.166 </details> --- ### Configuration 📅 **Schedule**: Branch creation - At any time (no schedule defined), Automerge - At any time (no schedule defined). 🚦 **Automerge**: Disabled by config. Please merge this manually once you are satisfied. ♻ **Rebasing**: Whenever PR becomes conflicted, or you tick the rebase/retry checkbox. 🔕 **Ignore**: Close this PR and you won't be reminded about this update again. --- - [ ] <!-- rebase-check -->If you want to rebase/retry this PR, check this box --- This PR has been generated by [Mend Renovate](https://www.mend.io/free-developer-tools/renovate/). View repository job log [here](https://developer.mend.io/github/autoblocksai/javascript-sdk). <!--renovate-debug:eyJjcmVhdGVkSW5WZXIiOiIzNy4zMS41IiwidXBkYXRlZEluVmVyIjoiMzcuMzEuNSIsInRhcmdldEJyYW5jaCI6Im1haW4ifQ==--> --------- Co-authored-by: renovate[bot] <29139614+renovate[bot]@users.noreply.github.com> Co-authored-by: Nicole White <[email protected]>
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Labels
auto:improvement
Medium size change to existing code to handle new use-cases
lgtm
PRs that are ready to be merged as-is
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
This simple change allows users to fully overwrite the prompts when creating a chat agent, and optimize them for their use case. The change makes the agent code much more generally useful. Previously, they could overwrite everything except for FORMAT_INSTRUCTIONS, which is essential to customization.