← 返回 Skills 市场
tangweigang-jpg

Chroma Vector Db

作者 Tang Weigang · GitHub ↗ · v0.1.0 · MIT-0
cross-platform ⚠ suspicious
76
总下载
0
收藏
0
当前安装
1
版本数
在 OpenClaw 中安装
/install chroma-vector-db
功能描述
Chroma 向量数据库：Rust 内核（v1.0.0+ 重写，2025-03），多语言客户端SDK。单节点用 PersistentClient（SQLite + 本地 HNSW）或 EphemeralClient（内存）；分布式 / 云用 SPANN + BLOCKFILE on S3/GCS。 Chroma...
使用说明 (SKILL.md)

这个 skill 适合什么用户？能做哪些任务？

概览

Chroma 是 Rust 内核的开源向量数据库（github.com/chroma-core/chroma，v1.0.0+ 2025-03 重写）。单节点模式用 PersistentClient（SQLite 元数据 + 本地 HNSW 索引）或 EphemeralClient（内存）；分布式 / 云模式用 SPANN 索引+ 自研 BLOCKFILE 存储 on S3/GCS。

数据路径：Client Factory → API 层（v1.0.0 起 Rust 默认）→ Segment 层（本地 2 段 / 分布式 3 段）→ Index（HNSW 或 SPANN）→ 持久化。Embed...

Doramagic 晶体页: https://doramagic.ai/zh/crystal/chroma-vector-db

知识规模

  • 52 条约束 (3 fatal + 49 non-fatal)
  • 上游源码: chroma-core/chroma @ commit 598f85f0
  • 蓝图 ID: finance-bp-138

用法

Host AI（Claude Code / Cursor / OpenClaw）读 references/seed.yaml，按其中的：

  • intent_router 匹配用户意图
  • architecture 理解项目架构
  • constraints 应用 anti-pattern 约束
  • business_decisions 参考核心设计决策

FAQ 摘要

这个 skill 适合什么用户？能做哪些任务？

适合需要向量数据库支持 RAG / 推荐 / 语义搜索的工程师：原型开发用EphemeralClient，本地小规模用 PersistentClient，生产分布式用云SPANN。25+ EmbeddingFunction 一键切换。访问 doramagic.ai/r/chroma 查看完整用例。

需要准备什么环境？依赖什么？

SQLite ≥ 3.35.0 是硬要求（chromadb/init.py:137-155 检查并 raise，Colab 自动 hot-swap 到 pysqlite3-binary）。Python ≥ 3.9。服务端 / 分布式需要 Rust 工具链或预构建 docker。可选：hosted EF 的 provider API key（OpenAI / Cohere 等）。

会踩哪些坑？这个 skill 怎么防护？

本 skill 内置 52 条约束（3 条 fatal）。典型踩坑：(1) cosine 实现实际是 1 - dot(a,b)，假定向量预归一化——未归一化数据score 会偏；(2) v1.0.0 Rust 默认静默忽略 4 个旧 Python 设置（chroma_server_nofile 等）——配置文件没报错但不生效；

完整文档: 见 references/seed.yaml (v6.1 schema). 浏览页: https://doramagic.ai/zh/crystal/chroma-vector-db

安全使用建议
This skill is instruction-only and sourced with no homepage; before installing consider: (1) It will ask the host agent to run Python/SQLite/zvt checks and touch files like ~/.zvt — run it only in an isolated environment or sandbox. (2) The skill manifest does not declare required binaries or env vars even though seed.yaml requires them — ask the publisher to provide a clear requirements list (python3 >=3.9, SQLite >=3.35, zvt, any provider API keys) and to confirm which host files will be read/written. (3) Seed.yaml may require reading LATEST.yaml / LATEST.jsonl; ensure those files don't contain secrets or sensitive configs before letting the agent access them. (4) If you need lower risk, request a version that documents required host commands and that can run in a container, or run the skill manually after inspecting seed.yaml and its referenced files. I have medium confidence due to the manifest/instruction mismatch; providing an explicit requirements manifest and confirmation that no external endpoints are contacted would raise confidence toward 'benign'.
功能分析
Type: OpenClaw Skill Name: chroma-vector-db Version: 0.1.0 The skill bundle exhibits a severe discrepancy between its stated purpose and its internal logic. While SKILL.md and _meta.json describe a Chroma Vector Database utility, the core logic in references/seed.yaml is almost entirely focused on quantitative financial trading using the 'zvt' library. The seed.yaml contains 'semantic locks' (SL-01 to SL-12) and constraints (chroma-C-*) that enforce financial trading rules such as 'sell before buy' ordering, A-share T+1 constraints, and specific MACD parameters. Additionally, the install_recipes and preconditions attempt to install and initialize the 'zvt' trading framework. This misleading configuration is highly irregular and could lead an AI agent to execute financial trading logic or request brokerage credentials (JoinQuant/QMT) under the guise of a database management task.
能力标签
cryptorequires-sensitive-credentials
能力评估
⚠ Purpose & Capability
SKILL.md/metadata present this as an AI-engineering knowledge skill that requires 'no installation' and declares no binaries or env vars, yet the included seed.yaml and preconditions clearly expect python3, SQLite (≥3.35.0), the zvt package, writable ZVT_HOME (~/.zvt), and potentially a Rust toolchain or Docker for server builds. These runtime needs are not declared in the skill manifest, which is inconsistent.
⚠ Instruction Scope
The instructions tell the host agent to re-read seed.yaml on any behavioral decision, run precondition shell checks (python -c 'import zvt', touch files, assert K-data present), and consult LATEST.yaml/LATEST.jsonl for traceback. Those steps involve executing host-side commands and reading/writing local filesystem paths — scope beyond a simple 'knowledge' skill and not limited to the stated purpose.
✓ Install Mechanism
No install spec or archive downloads are included (instruction-only). This minimizes supply-chain risk because nothing is automatically written or fetched by the skill itself.
⚠ Credentials
The manifest lists no required environment variables or credentials, yet instructions reference ZVT_HOME, require SQLite/Python, and mention optional hosted embedding provider API keys. The absence of declared env requirements while expecting to access local config and optional provider keys is disproportionate and inconsistent.
ℹ Persistence & Privilege
always:false and no install actions are declared (good). However, the seed.yaml's execution_protocol and workspace_resolution assume filesystem access and may cause repeated host-side checks; autonomous invocation is allowed (default) which increases blast radius if the host grants filesystem/command execution privileges. The skill does not request permanent inclusion or modify other skills.
如何使用
  1. 确保已安装 OpenClaw（本地或 Docker 部署）
  2. 在对话框中输入安装命令：/install chroma-vector-db
  3. 安装完成后，直接呼叫该 Skill 的名称或使用 /chroma-vector-db 触发
  4. 根据 Skill 的参数说明提供必要输入，即可获得结构化输出
版本历史
v0.1.0
Chroma vector database skill (Rust core) — 52 constraints / 3 fatal. Default cosine via EF, dim latch.
元数据
Slug chroma-vector-db
版本 0.1.0
许可证 MIT-0
累计安装 0
当前安装数 0
历史版本数 1
常见问题

Chroma Vector Db 是什么？

Chroma 向量数据库：Rust 内核（v1.0.0+ 重写，2025-03），多语言客户端SDK。单节点用 PersistentClient（SQLite + 本地 HNSW）或 EphemeralClient（内存）；分布式 / 云用 SPANN + BLOCKFILE on S3/GCS。 Chroma... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 76 次。

如何安装 Chroma Vector Db？

在 OpenClaw 或 Claude Code 对话框中运行命令「/install chroma-vector-db」即可一键安装，无需额外配置。

Chroma Vector Db 是免费的吗？

是的，Chroma Vector Db 完全免费，采用 MIT-0 许可证，可自由下载、安装和使用。

Chroma Vector Db 支持哪些平台？

Chroma Vector Db 跨平台运行，可在任意部署了 OpenClaw / Claude Code 的环境中使用（cross-platform）。

谁开发了 Chroma Vector Db？

由 Tang Weigang（@tangweigang-jpg）开发并维护，当前版本 v0.1.0。

推荐 Skills
self-improving agent
Captures learnings, errors, and corrections to enable continuous improvement. Use when: (1) A command or operation fails unexpectedly, (2) User corrects Clau...
⬇ 463283 · by pskoett
Skill Vetter
Security-first skill vetting for AI agents. Use before installing any skill from ClawdHub, GitHub, or other sources. Checks for red flags, permission scope, and suspicious patterns.
⬇ 259410 · by spclaudehome
Polymarket
Query Polymarket prediction markets. Check odds, find trending markets, search events, track price movements.
⬇ 232503 · by joelchance
Self-Improving + Proactive Agent
Self-reflection + Self-criticism + Self-learning + Self-organizing memory. Agent evaluates its own work, catches mistakes, and improves permanently. Use when...
⬇ 200423 · by ivangdavila
Github
Interact with GitHub using the `gh` CLI. Use `gh issue`, `gh pr`, `gh run`, and `gh api` for issues, PRs, CI runs, and advanced queries.
⬇ 191115 · by steipete
ontology
Typed knowledge graph for structured agent memory and composable skills. Use when creating/querying entities (Person, Project, Task, Event, Document), linkin...
⬇ 190156 · by oswalpalash

💬 留言讨论