vector-indexing
Build and troubleshoot DocArray Document Index workflows for local exact search, filters, hybrid queries, persistence, nested subindexes, and optional vector databases.
- License 许可
- Apache-2.0
- 最近更新
- 2026-10-08
- Stars
- ★ 11
- Forks
- 3
01 前言 Front Matter
- name
- vector-indexing
- license
- Apache 2.0
- metadata
- { "disco-role": "operating" }
- description
- Build and troubleshoot DocArray Document Index workflows for local exact search, filters, hybrid queries, persistence, nested subindexes, and optional vector databases.
- disable-model-invocation
- true
02 资源文件 4
- 参考文档references/api-reference.md4.0 KB
- 参考文档references/index-workflows.md4.8 KB
- 参考文档references/optional-backends.md3.8 KB
- 参考文档references/troubleshooting.md4.5 KB
03 SKILL.md 正文
DocArray vector indexing
Use this sub-skill when a task needs to index typed documents, search embeddings, filter metadata, combine queries, persist a local index, or select an optional vector database backend.
Route here for
- Defining a dimensioned vector field for
InMemoryExactNNIndexor anotherDocumentIndex. - Indexing
DocListdata and usingfind(),find_batched(),filter(),filter_batched(), ID access, deletion, or re-index updates. - Building local hybrid queries with
build_query()andexecute_query(). - Searching nested
DocListfields with a subindex. - Choosing HNSWLib, Qdrant, Weaviate, Elasticsearch, Redis, Milvus, MongoDB Atlas, or Epsilla integrations and diagnosing missing extras or services.
Route elsewhere
- Base schemas, predefined documents, tensor typing, or
DocList/DocVecmodeling before indexing: use siblingdocument-modeling. - JSON/protobuf/bytes/CSV/DataFrame serialization or
file:///S3 storage: use siblingserialization-storage.
Read these bundled references
- Start with API reference for verified index signatures, schema requirements, and configuration fields.
- Use Index workflows for local indexing, search, filters, query builders, persistence, and nested subindexes.
- Use Optional backends before installing an external backend or starting a service.
- Use Troubleshooting for schema, dimension, query-builder, dependency, and service failures.
Safe bundled helper
Run inmemory_index_smoke.py for deterministic CPU-only local checks. It uses only NumPy and the in-memory index; it does not start a database, download data, or require credentials:
Run these commands from this sub-skill root (the directory containing this SKILL.md):
python scripts/inmemory_index_smoke.py --help
python scripts/inmemory_index_smoke.py
python scripts/inmemory_index_smoke.py --exercise-persist
The verified minimum scope is CPU with the in-memory backend. External services, HNSWLib, and accelerator-specific tensor paths remain optional and unverified until a task prepares their dependencies and service/hardware plan.
04 同仓库更多 Skills
obsidian-clipper-template-creator
Guide for creating templates for the Obsidian Web Clipper. Use when you want to create a new clipping template, understand available variables, or format clipped content.
obsidian-clipper-template-creator
Guide for creating templates for the Obsidian Web Clipper. Use when you want to create a new clipping template, understand available variables, or format clipped content.
python-api
Programmatic Object-Detection-Metrics API guidance for BoundingBox, BoundingBoxes, Evaluator, enums, coordinate conversion, AP/mAP semantics, and source-style import diagnostics.
file-evaluation
Folder-based PASCAL VOC AP/mAP evaluation from ground-truth and detection text files.
object-detection-metrics
Guides legacy Object-Detection-Metrics PASCAL VOC AP/mAP evaluation from detection text folders and direct Python API classes.
nx-workspace-patterns
Configure and optimize Nx monorepo workspaces. Use when setting up Nx, configuring project boundaries, optimizing build caching, or implementing affected commands.