{"id":15487850,"url":"https://github.com/werediver/qas","last_synced_at":"2025-03-28T16:16:10.213Z","repository":{"id":214100873,"uuid":"734469368","full_name":"werediver/qas","owner":"werediver","description":"A retrieval-augmented question answering system","archived":false,"fork":false,"pushed_at":"2024-02-18T16:18:49.000Z","size":95,"stargazers_count":0,"open_issues_count":0,"forks_count":0,"subscribers_count":3,"default_branch":"main","last_synced_at":"2024-10-19T08:13:37.853Z","etag":null,"topics":[],"latest_commit_sha":null,"homepage":null,"language":"Python","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"mit","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/werediver.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":"LICENSE","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null}},"created_at":"2023-12-21T19:02:46.000Z","updated_at":"2023-12-21T19:11:22.000Z","dependencies_parsed_at":"2023-12-28T13:56:34.523Z","dependency_job_id":"94812ca7-11b5-47c7-b6e2-644659d944c4","html_url":"https://github.com/werediver/qas","commit_stats":{"total_commits":19,"total_committers":1,"mean_commits":19.0,"dds":0.0,"last_synced_commit":"8b77f7a5b49e78daa53e83def8854ee73080d903"},"previous_names":["werediver/qas"],"tags_count":0,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/werediver%2Fqas","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/werediver%2Fqas/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/werediver%2Fqas/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/werediver%2Fqas/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/werediver","download_url":"https://codeload.github.com/werediver/qas/tar.gz/refs/heads/main","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":246059336,"owners_count":20717085,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2022-07-04T15:15:14.044Z","host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":[],"created_at":"2024-10-02T06:41:53.795Z","updated_at":"2025-03-28T16:16:10.189Z","avatar_url":"https://github.com/werediver.png","language":"Python","funding_links":[],"categories":[],"sub_categories":[],"readme":"# A retrieval-augmented question-answering system\n\nThe goal of this project is to implement a retrieval-augmented (RAG) question-answering system that uses local documents or a remote wiki-like system (e.g. Confluence) to augment the user requests with relevant context before passing them to an LLM.\n\nThe current implementation relies on [Ollama](https://github.com/jmorganca/ollama) for text generation and [fastembed](https://github.com/qdrant/fastembed) / [BAAI/bge-small-en-v1.5](https://huggingface.co/BAAI/bge-small-en-v1.5) for text embedding.\n\nUsing an OpenAI API-like provider instead of Ollama (e.g. [LM Studio](https://lmstudio.ai/)) is easily possible.\n\n## How to run\n\nYou'll need [PDM](https://github.com/pdm-project/pdm) and Ollama installed in your system.\n\nFirst, install the project dependencies by executing the following command in the project directory:\n\n```\npdm install\n```\n\nThen make sure Ollama has Mistral model downloaded by executing the following command:\n\n```\nollama pull mistral\n```\n\nMake sure Ollama server is running by executing the following command or in any other way:\n\n```\nollama serve\n```\n\nFinally, to run the app execute the following command:\n\n```\nenv DOCS=\"path/to/txt/or/md/docs\" pdm run src/app.py\n```\n\n## How to load Confluence pages\n\n`confluence_md` package can be used as a CLI tool to download Confluence pages as Markdown files with metadata in stored YAML front matter.\n\nMake sure to set the following environment variables put them in `.env` file in the project root:\n\n- `URL`, Confluence server URL\n- `CLIENT_ID`, Confluence user name\n- `ACCESS_TOKEN`, personal access token\n- `DUMP_DIR`, directory to write Markdown files to\n\nTo start downloading, execute the following command:\n\n```\ndate; time pdm run python -m src.confluence_md\n```\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fwerediver%2Fqas","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fwerediver%2Fqas","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fwerediver%2Fqas/lists"}