- New data loader - Opensearch (#672)
- AddMainContentExtractorReader (#679)
- Image reader was ignoring metadata. Added metadata to loaded ImageDocument (#668)
- Add Cypher validation to Neo4j db tool (#660)
- feat: Improvements to Chroma loader (#673)
- Adding Earning Call transcripts of US based companies (#658)
- add amazon product extraction (#670)
- Add Waii connector (#647)
- Citation LlamaPack (#666)
- Add some advanced retrieval llama packs (#659)
- Add Llama Packs (#646)
- Unstructured.IO API support (#648)
- Updated init file (#633)
- fIMDB Movie reviews bug fixes and feature addition (#636)
- replace s/gpt_index/llama_index references in READMEs (#635)
- Extend GoogleSheetsReader to Accept Custom Text Column Names (#620)
- feat: sync mongo to SimpleMongoReader of llama-index (#624)
- Adding moview reviews link for IMDB loader (#630)
- Update tavily requirements (#618)
- fix tavily tool (#617)
- Add a Lilac dataset reader. (#563)
- Cogniswitch llama-hub corrections (#613)
- name change and README update to Tavily (#614)
- CogniSwitch Connector (#604)
- add docs to min_chunk_size (#607)
- Change Tavily client name (#609)
- Added OpenAlex Reader for Scientific QA (#599)
- Added Tavily Search API as a tool (#600)
- Adding loader to read from OneDrive Personal and OneDrive for Business (#597)
- Update TrafilaturaWebReader in library.json (#602)
- Added smart layout aware fast PDF reader/loader (#587)
- Added Protein Data Bank citation reader (#595)
- Fix Firestore client info (#586)
- doc(README): remove typo (#589)
- Add model kwargs to image loader (#588)
- updating documentation to match method defaults (#591)
- doc(README): Fix Grammatical Error (#593)
- Fix import statement for CodeInterpreterToolSpec (#594)
- IMDB Review loader (#571)
- Add AWS Athena Reader (#579)
- add PatentsviewReader for reading patent abstract (#582)
- Add proper init.py files (#581)
- Add possibility to pass model kwargs to image loader models
- Add Nougat OCR loader (#541)
- improve bitbucket loader and extension to skip (#576)
- Add RayyanLoader to fetch review articles from Rayyan (#570)
- added bitbucket loader (#572)
- fix: credentials path and readme improvement (#567)
- fix: pdf google drive (#568)
- Updating URLs in Wikipedia Loader (#569)
- Loader for Macrometa GDN (#484)
- adding boto3 minio doc loader (#497)
- Add new data reader: AssemblyAIAudioTranscriptReader (#562)
- fix: PyMuPDF Reader broken (#547)
- Add page id to extra_info (#542)
- Update: Readme with corrected example url for Playgrounds_subgraph_connector tool (#551)
- add url and status to confluence loader document (#553)
- Changes from llama_index/PR#7906 (#557)
- fix: sql_wrapper utilities (#558)
- Adding functionality for AsanaReader (#386)
- Add JSONL functionality to JSONReader (#552)
- add url and timestamp to slack loader documents metadata (#559)
- add url to asana loader docs (#560)
- Added oAuth to Jira loader (#272)
- refactor: add custom path for download_loader to enable functionality (#318)
- feat: Add PlaygroundsSubgraphInspectorToolSpec to llama_hub (#535)
- add full formatting and linting (#537, #538)
- Add new data reader: AssemblyAIAudioTranscriptReader (#562)
- fix: added missing loaders removed from a old PR (#540)
- break loop if API error for slack reader (#544)
- mbox: allow custom, stable document id (#393)
- update database doc (#531)
- fix: jsondata loader on library (#539)
- fix: remove isort due using black and add a new checklist (#546)
- Neo4j Schema Query Builder Integration (#520)
- feat: Add PlaygroundsSubgraphConnector to Llama Hub (#528)
- Fix BaseGithubClient and _generate_documents (#526)
- Refactor load_data for consistent sheet handling and cleaner code (#488)
- remove redundant if/else from imports in github readers (#524)
- fix: 🚑️ remove app id and secret from feishu reader (#525)
- Add reader for GitHub collaborators (#512)
- HWPReader (#517)
- fixed typos in the readme.md of salesforce tool (#515)
- Service account support for google drive loader (#513)
- Enhance PDFReader to accept File object as well, in addition to a path string (#514)
- add urls to metadata saved by github repo reader (#522)
- Elastic data reader (#508)
- Salesforce Tool (#507)
- add HWPReader to JSON (#500)
- Add issue labels to
extra_info
saved by GithubRepositoryIssuesReader (#502)
- Adding MultiOn browsing tool (#481)
- fix olefile import
- add hanguel / hwp readers (#493)
- Enhancement to Semantic Scholar Loader - full text reader (#482)
- Adding hierarchical agent example and comparison (#495)
- fix transforming error in wordlift reader (#501)
(includes v0.0.25)
- Add Linear loader (#490)
- Add PDF Table Reader (#476)
- Bagel loader Added (#479)
- Database arg fix in Firestore client (#483)
- Some update to prevent errors when transforming data in wordlift loader (#489)
- UTF-8 encode and decode for gmail (#491)
- iterate json data to Document object in unstructured loader (#485)
- add custom user agent for metaphor llama index initialization (#480)
- Fix Syntax in Docs (#478)
- Adding Metaphor tool and notebook (#466)
- Adding ArXiv tool spec (#464)
- Adding Azure speech, translate and CV tools (#459)
- SDLReader for Graphql (#461)
- missing import added for wikipedia (#463)
- patch document in wordpress (#462)
- ZepReader (#452)
- GraphQL Tool Spec (#455)
- Adding PythonFileToolSpec (#453)
- Adding bing search tools (#457)
- RSS Feed / OPML reader and article parser (#444)
- Adding Shopify GraphQL Tool Spec and Demo (#442)
- changed num to init for better declaration in google search tool (#449)
- added a num parameter to the google_search (#446)
- Added Semantic scholar reader (#439)
- Update docugami loader notebook (#445)
- Remove release workflow in favor of using tags (#443)
- Auto-Tool creation notebook (#424)
- Security warning for Code Interpreter and lock requests headers to a domain (#438)
- A few more tags based on GA (#437)
- Add publish release workflow (#436)
- add retrieval augmented text-to-image example (#434)
- hatena blog reader add url (#425)
- adding more functions to DeepDoctectionReader and docs (#431)
- Gurureader (#427)
- feat: minio loader (#430)
- Adding SEC Filings Loader (#415)
- Adding some tags for llama hub searches (#422)
- Update unstructured loader README (#418)
- synced with llama_index/readers/file/markdown_reader.py (#388)
- YoutubeTranscriptReader tests (#412)
- fix some bugs in WordLift loader (#421)
- Adding ChatGPT plugin tool (#405)
- Adding Dalle2 text to image tool (#407)
- Add SingleStoreReader (#404)
- Update GmailReader to return internalDate (#406)
- Update ChromaReader to use 0.4.0 API (#394)
- Update Zapier to expose a list of tools (#401)
- Add code interpreter tool (#398)
- Add Feishu Docs Reader (#383)
- Add Google Keep reader (#370)
- Various bug fix and improvements to pandas excel reader (#397, #372, #391)
- Update README.md to better highlight data agents (#395)
- Update Zapier tool to use parameters in addition to instructions (#390)
- Make S3Reader more configurable (#364)
- Add weather agent tool (#384)
- fix tool readme imports (#381)
- fix tool notebooks (#380)
- fix gmail notebook (#379)
- Adding Agent Tools to LlamaHub (#377)
- Docugami: use metadata over deprecated extra_info (#375)
- change readme and requirements (#354)
- Update zendesk loader (#358)
- Remove ascii in gmail loader (#361)
- Fix readme for wordlift (#357)
- address bug in paginated loader that always fetches the next page, even if the max_results is met (#363)
- S3 Extensions Filtering (#362)
- Add argument encoding to specify the encoding to open the csv file. (#367)
- Update README.md for unstructured (#366)
- fix bug where messages are returned without getting message data (#368)
- Add firebase realtime db (#347)
- Add WordLift Loader (#346)
- [bugfix] replace None to empty string to avoid TypeError (#351)
- chore: add youtube_id to document metadata (#352)
- Add extra_info with Source to web reader (#337)
- support pagination on gmail loader (#348)
- feat: add source metadata to documents created by the GH issues reader (#341)
- Update markdown reader (#344)
- ensure metadata always valid (#343)
- update document usage (#338). NOTE: all Document objects must be prefixed with a
kwargs
field.
- feat: add document reader for GH issues (#332)
- Add sitemap loader (#328)
- Fixed ValueError: Metadata key must be str! error (#317)
- update namespace (#324)
- add changelog (#333)
- Fix KibelaReader (#334)
- confluence.load_data new features, bug fixes, tests (#330)
- Update readme files for reference (#336)
- Add KibelaReader (#319)
- reader: add deplot tabular graph image reader (#321)
- support Confluence personal access token (#323)
- service account credentials (#316)
- None
- None