# postgresml - Doramagic AI Context Pack > 定位：安装前体验与判断资产。它帮助宿主 AI 有一个好的开始，但不代表已经安装、执行或验证目标项目。 ## 充分原则 - **充分原则，不是压缩原则**：AI Context Pack 应该充分到让宿主 AI 在开工前理解项目价值、能力边界、使用入口、风险和证据来源；它可以分层组织，但不以最短摘要为目标。 - **压缩策略**：只压缩噪声和重复内容，不压缩会影响判断和开工质量的上下文。 ## 给宿主 AI 的使用方式 你正在读取 Doramagic 为 postgresml 编译的 AI Context Pack。请把它当作开工前上下文：帮助用户理解适合谁、能做什么、如何开始、哪些必须安装后验证、风险在哪里。不要声称你已经安装、运行或执行了目标项目。 ## Claim 消费规则 - **事实来源**：Repo Evidence + Claim/Evidence Graph；Human Wiki 只提供显著性、术语和叙事结构。 - **事实最低状态**：`supported` - `supported`：可以作为项目事实使用，但回答中必须引用 claim_id 和证据路径。 - `weak`：只能作为低置信度线索，必须要求用户继续核实。 - `inferred`：只能用于风险提示或待确认问题，不能包装成项目事实。 - `unverified`：不得作为事实使用，应明确说证据不足。 - `contradicted`：必须展示冲突来源，不得替用户强行选择一个版本。 ## 它最适合谁 - **想在安装前理解开源项目价值和边界的用户**：当前证据主要来自项目文档。 证据：`README.md` Claim：`clm_0002` supported 0.86 ## 它能做什么 - **命令行启动或安装流程**（需要安装后验证）：项目文档中存在可执行命令，真实使用需要在本地或宿主环境中运行这些命令。 证据：`packages/pgml-rds-proxy/download-pgcat.sh`, `pgml-apps/pgml-chat/README.md`, `pgml-cms/blog/announcing-gptq-and-ggml-quantized-llm-support-for-huggingface-transformers.md`, `pgml-cms/blog/pgml-chat-a-command-line-tool-for-deploying-low-latency-knowledge-based-chatbots-part-i.md` 等 Claim：`clm_0001` supported 0.86 ## 怎么开始 - `pip install -r requirements.txt` 证据：`pgml-cms/blog/announcing-gptq-and-ggml-quantized-llm-support-for-huggingface-transformers.md` Claim：`clm_0003` unverified 0.25 - `pip install pgml-chat` 证据：`pgml-cms/blog/pgml-chat-a-command-line-tool-for-deploying-low-latency-knowledge-based-chatbots-part-i.md` Claim：`clm_0004` unverified 0.25 - `git clone https://github.com/postgresml/Auto-GPT` 证据：`pgml-cms/blog/postgresml-as-a-memory-backend-to-auto-gpt.md` Claim：`clm_0005` unverified 0.25 - `curl \` 证据：`pgml-cms/blog/using-postgresml-with-django-and-embedding-search.md` Claim：`clm_0006` unverified 0.25, `clm_0011` unverified 0.25, `clm_0022` supported 0.86, `clm_0024` supported 0.86 等 - `npm i pgml` 证据：`pgml-cms/docs/introduction/getting-started/connect-your-app.md` Claim：`clm_0007` supported 0.86 - `pip install pgml` 证据：`pgml-cms/docs/introduction/getting-started/connect-your-app.md` Claim：`clm_0004` unverified 0.25, `clm_0008` supported 0.86 - `git clone https://github.com/postgresml/postgresml && \` 证据：`pgml-cms/docs/open-source/pgml/developers/self-hosting/building-from-source.md` Claim：`clm_0009` unverified 0.25, `clm_0012` unverified 0.25 - `git clone https://github.com/postgresml/pgcat` 证据：`pgml-cms/docs/open-source/pgml/developers/self-hosting/pooler.md` Claim：`clm_0010` unverified 0.25 - `curl --proto '=https' --tlsv1.2 -sSf https://sh.rustup.rs | sh` 证据：`pgml-cms/docs/open-source/pgml/developers/self-hosting/pooler.md` Claim：`clm_0011` unverified 0.25 - `git clone https://github.com/postgresml/postgresml` 证据：`pgml-cms/docs/open-source/pgml/developers/installation.md` Claim：`clm_0009` unverified 0.25, `clm_0012` unverified 0.25 ## 继续前判断卡 - **当前建议**：先做角色匹配试用 - **为什么**：这个项目更像角色库，核心风险是选错角色或把角色文案当执行能力；先用 Prompt Preview 试角色匹配，再决定是否沙盒导入。 ### 30 秒判断 - **现在怎么做**：先做角色匹配试用 - **最小安全下一步**：先用 Prompt Preview 试角色匹配；满意后再隔离导入 - **先别相信**：角色质量和任务匹配不能直接相信。 - **继续会触碰**：角色选择偏差、命令执行、本地环境或项目文件 ### 现在可以相信 - **适合人群线索：想在安装前理解开源项目价值和边界的用户**（supported）：有 supported claim 或项目证据支撑，但仍不等于真实安装效果。 证据：`README.md` Claim：`clm_0002` supported 0.86 - **能力存在：命令行启动或安装流程**（supported）：可以相信项目包含这类能力线索；是否适合你的具体任务仍要试用或安装后验证。 证据：`packages/pgml-rds-proxy/download-pgcat.sh`, `pgml-apps/pgml-chat/README.md`, `pgml-cms/blog/announcing-gptq-and-ggml-quantized-llm-support-for-huggingface-transformers.md`, `pgml-cms/blog/pgml-chat-a-command-line-tool-for-deploying-low-latency-knowledge-based-chatbots-part-i.md` 等 Claim：`clm_0001` supported 0.86 - **存在 Quick Start / 安装命令线索**（supported）：可以相信项目文档出现过启动或安装入口；不要因此直接在主力环境运行。 证据：`pgml-cms/blog/using-postgresml-with-django-and-embedding-search.md` Claim：`clm_0006` unverified 0.25, `clm_0011` unverified 0.25, `clm_0022` supported 0.86, `clm_0024` supported 0.86 ### 现在还不能相信 - **角色质量和任务匹配不能直接相信。**（unverified）：角色库证明有很多角色，不证明每个角色都适合你的具体任务，也不证明角色能产生高质量结果。 - **不能把角色文案当成真实执行能力。**（unverified）：安装前只能判断角色描述和任务画像是否匹配，不能证明它能在宿主 AI 里完成任务。 - **真实输出质量不能在安装前相信。**（unverified）：Prompt Preview 只能展示引导方式，不能证明真实项目中的结果质量。 - **宿主 AI 版本兼容性不能在安装前相信。**（unverified）：Claude、Cursor、Codex、Gemini 等宿主加载规则和版本差异必须在真实环境验证。 - **不会污染现有宿主 AI 行为，不能直接相信。**（inferred）：Skill、plugin、AGENTS/CLAUDE/GEMINI 指令可能改变宿主 AI 的默认行为。 - **可安全回滚不能默认相信。**（unverified）：除非项目明确提供卸载和恢复说明，否则必须先在隔离环境验证。 - **真实安装后是否与用户当前宿主 AI 版本兼容？**（unverified）：兼容性只能通过实际宿主环境验证。 - **项目输出质量是否满足用户具体任务？**（unverified）：安装前预览只能展示流程和边界，不能替代真实评测。 ### 继续会触碰什么 - **角色选择偏差**：用户对任务应该由哪个专家角色处理的判断。 原因：选错角色会让 AI 从错误专业视角回答，浪费时间或误导决策。 - **命令执行**：包管理器、网络下载、本地插件目录、项目配置或用户主目录。 原因：运行第一条命令就可能产生环境改动；必须先判断是否值得跑。 证据：`packages/pgml-rds-proxy/download-pgcat.sh`, `pgml-apps/pgml-chat/README.md`, `pgml-cms/blog/announcing-gptq-and-ggml-quantized-llm-support-for-huggingface-transformers.md`, `pgml-cms/blog/pgml-chat-a-command-line-tool-for-deploying-low-latency-knowledge-based-chatbots-part-i.md` 等 - **本地环境或项目文件**：安装结果、插件缓存、项目配置或本地依赖目录。 原因：安装前无法证明写入范围和回滚方式，需要隔离验证。 证据：`packages/pgml-rds-proxy/download-pgcat.sh`, `pgml-apps/pgml-chat/README.md`, `pgml-cms/blog/announcing-gptq-and-ggml-quantized-llm-support-for-huggingface-transformers.md`, `pgml-cms/blog/pgml-chat-a-command-line-tool-for-deploying-low-latency-knowledge-based-chatbots-part-i.md` 等 - **宿主 AI 上下文**：AI Context Pack、Prompt Preview、Skill 路由、风险规则和项目事实。 原因：导入上下文会影响宿主 AI 后续判断，必须避免把未验证项包装成事实。 ### 最小安全下一步 - **先跑 Prompt Preview**：先用交互式试用验证任务画像和角色匹配，不要先导入整套角色库。（适用：任何项目都适用，尤其是输出质量未知时。） - **只在隔离目录或测试账号试装**：避免安装命令污染主力宿主 AI、真实项目或用户主目录。（适用：存在命令执行、插件配置或本地写入线索时。） - **安装后只验证一个最小任务**：先验证加载、兼容、输出质量和回滚，再决定是否深用。（适用：准备从试用进入真实工作流时。） ### 退出方式 - **保留安装前状态**：记录原始宿主配置和项目状态，后续才能判断是否可恢复。 - **保留原始角色选择记录**：如果输出偏题，可以回到任务画像阶段重新选择角色，而不是继续沿着错误角色推进。 - **记录安装命令和写入路径**：没有明确卸载说明时，至少要知道哪些目录或配置需要手动清理。 - **如果没有回滚路径，不进入主力环境**：不可回滚是继续前阻断项，不应靠信任或运气继续。 ## 哪些只能预览 - 解释项目适合谁和能做什么 - 基于项目文档演示典型对话流程 - 帮助用户判断是否值得安装或继续研究 ## 哪些必须安装后验证 - 真实安装 Skill、插件或 CLI - 执行脚本、修改本地文件或访问外部服务 - 验证真实输出质量、性能和兼容性 ## 边界与风险判断卡 - **把安装前预览误认为真实运行**：用户可能高估项目已经完成的配置、权限和兼容性验证。 处理方式：明确区分 prompt_preview_can_do 与 runtime_required。 Claim：`clm_0027` inferred 0.45 - **命令执行会修改本地环境**：安装命令可能写入用户主目录、宿主插件目录或项目配置。 处理方式：先在隔离环境或测试账号中运行。 证据：`packages/pgml-rds-proxy/download-pgcat.sh`, `pgml-apps/pgml-chat/README.md`, `pgml-cms/blog/announcing-gptq-and-ggml-quantized-llm-support-for-huggingface-transformers.md`, `pgml-cms/blog/pgml-chat-a-command-line-tool-for-deploying-low-latency-knowledge-based-chatbots-part-i.md` 等 Claim：`clm_0028` supported 0.86 - **待确认**：真实安装后是否与用户当前宿主 AI 版本兼容？。原因：兼容性只能通过实际宿主环境验证。 - **待确认**：项目输出质量是否满足用户具体任务？。原因：安装前预览只能展示流程和边界，不能替代真实评测。 - **待确认**：安装命令是否需要网络、权限或全局写入？。原因：这影响企业环境和个人环境的安装风险。 ## 开工前工作上下文 ### 加载顺序 - 先读取 how_to_use.host_ai_instruction，建立安装前判断资产的边界。 - 读取 claim_graph_summary，确认事实来自 Claim/Evidence Graph，而不是 Human Wiki 叙事。 - 再读取 intended_users、capabilities 和 quick_start_candidates，判断用户是否匹配。 - 需要执行具体任务时，优先查 role_skill_index，再查 evidence_index。 - 遇到真实安装、文件修改、网络访问、性能或兼容性问题时，转入 risk_card 和 boundaries.runtime_required。 ### 任务路由 - **命令行启动或安装流程**：先说明这是安装后验证能力，再给出安装前检查清单。 边界：必须真实安装或运行后验证。 证据：`packages/pgml-rds-proxy/download-pgcat.sh`, `pgml-apps/pgml-chat/README.md`, `pgml-cms/blog/announcing-gptq-and-ggml-quantized-llm-support-for-huggingface-transformers.md`, `pgml-cms/blog/pgml-chat-a-command-line-tool-for-deploying-low-latency-knowledge-based-chatbots-part-i.md` 等 Claim：`clm_0001` supported 0.86 ### 上下文规模 - 文件总数：1213 - 重要文件覆盖：40/1213 - 证据索引条目：79 - 角色 / Skill 条目：73 ### 证据不足时的处理 - **missing_evidence**：说明证据不足，要求用户提供目标文件、README 段落或安装后验证记录；不要补全事实。 - **out_of_scope_request**：说明该任务超出当前 AI Context Pack 证据范围，并建议用户先查看 Human Manual 或真实安装后验证。 - **runtime_request**：给出安装前检查清单和命令来源，但不要替用户执行命令或声称已执行。 - **source_conflict**：同时展示冲突来源，标记为待核实，不要强行选择一个版本。 ## Prompt Recipes ### 适配判断 - 目标：判断这个项目是否适合用户当前任务。 - 预期输出：适配结论、关键理由、证据引用、安装前可预览内容、必须安装后验证内容、下一步建议。 ```text 请基于 postgresml 的 AI Context Pack，先问我 3 个必要问题，然后判断它是否适合我的任务。回答必须包含：适合谁、能做什么、不能做什么、是否值得安装、证据来自哪里。所有项目事实必须引用 evidence_refs、source_paths 或 claim_id。 ``` ### 安装前体验 - 目标：让用户在安装前感受核心工作流，同时避免把预览包装成真实能力或营销承诺。 - 预期输出：一段带边界标签的体验剧本、安装后验证清单和谨慎建议；不含真实运行承诺或强营销表述。 ```text 请把 postgresml 当作安装前体验资产，而不是已安装工具或真实运行环境。 请严格输出四段： 1. 先问我 3 个必要问题。 2. 给出一段“体验剧本”：用 [安装前可预览]、[必须安装后验证]、[证据不足] 三种标签展示它可能如何引导工作流。 3. 给出安装后验证清单：列出哪些能力只有真实安装、真实宿主加载、真实项目运行后才能确认。 4. 给出谨慎建议：只能说“值得继续研究/试装”“先补充信息后再判断”或“不建议继续”，不得替项目背书。 硬性边界： - 不要声称已经安装、运行、执行测试、修改文件或产生真实结果。 - 不要写“自动适配”“确保通过”“完美适配”“强烈建议安装”等承诺性表达。 - 如果描述安装后的工作方式，必须使用“如果安装成功且宿主正确加载 Skill，它可能会……”这种条件句。 - 体验剧本只能写成“示例台词/假设流程”：使用“可能会询问/可能会建议/可能会展示”，不要写“已写入、已生成、已通过、正在运行、正在生成”。 - Prompt Preview 不负责给安装命令；如用户准备试装，只能提示先阅读 Quick Start 和 Risk Card，并在隔离环境验证。 - 所有项目事实必须来自 supported claim、evidence_refs 或 source_paths；inferred/unverified 只能作风险或待确认项。 ``` ### 角色 / Skill 选择 - 目标：从项目里的角色或 Skill 中挑选最匹配的资产。 - 预期输出：候选角色或 Skill 列表，每项包含适用场景、证据路径、风险边界和是否需要安装后验证。 ```text 请读取 role_skill_index，根据我的目标任务推荐 3-5 个最相关的角色或 Skill。每个推荐都要说明适用场景、可能输出、风险边界和 evidence_refs。 ``` ### 风险预检 - 目标：安装或引入前识别环境、权限、规则冲突和质量风险。 - 预期输出：环境、权限、依赖、许可、宿主冲突、质量风险和未知项的检查清单。 ```text 请基于 risk_card、boundaries 和 quick_start_candidates，给我一份安装前风险预检清单。不要替我执行命令，只说明我应该检查什么、为什么检查、失败会有什么影响。 ``` ### 宿主 AI 开工指令 - 目标：把项目上下文转成一次对话开始前的宿主 AI 指令。 - 预期输出：一段边界明确、证据引用明确、适合复制给宿主 AI 的开工前指令。 ```text 请基于 postgresml 的 AI Context Pack，生成一段我可以粘贴给宿主 AI 的开工前指令。这段指令必须遵守 not_runtime=true，不能声称项目已经安装、运行或产生真实结果。 ``` ## 角色 / Skill 索引 - 共索引 73 个角色 / Skill / 项目文档条目。 - **Overview**（project_doc）：The key concepts that make up PostgresML. 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`pgml-cms/docs/README.md` - **PostgresML architecture**（project_doc）：PostgresML is an extension for the PostgreSQL database server. It operates inside the database, using the same hardware to perform machine learning tasks. 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`pgml-cms/docs/TODO/architecture/README.md` - **Enterprise**（project_doc）：Enterprise plans are ideal large companies that have special compliance needs and deployment configurations; with options for cloud-prem VPC , on-prem, ACL’s and more. 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`pgml-cms/docs/cloud/enterprise/README.md` - **Getting started**（project_doc）：Getting starting with PostgresML, a GPU powered machine learning database. 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`pgml-cms/docs/introduction/getting-started/README.md` - **Import your data**（project_doc）：Import your data into PostgresML using one of many supported methods. 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`pgml-cms/docs/introduction/import-your-data/README.md` - **Logical replication**（project_doc）：Stream data from your primary database to PostgresML in real time using logical replication. 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`pgml-cms/docs/introduction/import-your-data/logical-replication/README.md` - **Tabular data**（project_doc）：Tabular data is data stored in tables. A table is a format that defines rows and columns, and is the most common type of data organization. Examples of tabular data are spreadsheets, database tables, CSV files, and Pandas dataframes. 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`pgml-cms/docs/introduction/import-your-data/storage-and-retrieval/README.md` - **Korvus**（project_doc）：Korvus is an SDK for JavaScript, Python and Rust implements common use cases and PostgresML connection management. 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`pgml-cms/docs/open-source/korvus/README.md` - **API**（project_doc）：PostgresML client SDK for JavaScript, Python and Rust API. 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`pgml-cms/docs/open-source/korvus/api/README.md` - **Example Applications**（project_doc）：PostgresML client SDK for JavaScript, Python and Rust implements common example apps. 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`pgml-cms/docs/open-source/korvus/example-apps/README.md` - **Guides**（project_doc）：PostgresML client SDK for JavaScript, Python and Rust guides for more complex uses. 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`pgml-cms/docs/open-source/korvus/guides/README.md` - **PgCat pooler**（project_doc）：PgCat, the PostgreSQL connection pooler and proxy with support for sharding, load balancing, failover, and many more features. 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`pgml-cms/docs/open-source/pgcat/README.md` - **SQL extension**（project_doc）：- 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`pgml-cms/docs/open-source/pgml/README.md` - **PGML API**（project_doc）：The pgml extension API. 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`pgml-cms/docs/open-source/pgml/api/README.md` - **pgml.predict**（project_doc）：- 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`pgml-cms/docs/open-source/pgml/api/pgml.predict/README.md` - **Developers**（project_doc）：Documentation relevant to self-hosting, compiling or contributing to PostgresML 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`pgml-cms/docs/open-source/pgml/developers/README.md` - **Self-hosting**（project_doc）：PostgresML is a Postgres extension, so running it is very similar to running a self-hosted PostgreSQL database server. A typical architecture consists of a primary database that will serve reads and writes, optional replicas to scale reads horizontally, and a pooler to load balance connections. 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`pgml-cms/docs/open-source/pgml/developers/self-hosting/README.md` - **Guides**（project_doc）：Long form examples demonstrating use cases for PostgresML 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`pgml-cms/docs/open-source/pgml/guides/README.md` - **Chatbots**（project_doc）：- 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`pgml-cms/docs/open-source/pgml/guides/chatbots/README.md` - **Embeddings**（project_doc）：Embeddings are a key building block with many applications in modern AI/ML systems. They are particularly valuable for handling various types of unstructured data like text, images, and more, providing a pathway to richer insights and improved performance. A common use case for embeddings is to provide semantic search capabilities that go beyond traditional keyword matching to the underlying meaning in the data. 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`pgml-cms/docs/open-source/pgml/guides/embeddings/README.md` - **LLMs**（project_doc）：PostgresML integrates 🤗 Hugging Face Transformers https://huggingface.co/transformers to bring state-of-the-art models into the data layer. There are tens of thousands of pre-trained models with pipelines to turn raw inputs into useful results. Many state of the art deep learning architectures have been published and made available for download. You will want to browse all the models https://huggingface.co/models av… 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`pgml-cms/docs/open-source/pgml/guides/llms/README.md` - **Supervised Learning**（project_doc）：A machine learning approach that uses labeled data 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`pgml-cms/docs/open-source/pgml/guides/supervised-learning/README.md` - **Architecture**（project_doc）：Postgres + GPUs for ML/AI applications. 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`README.md` - **Packages**（project_doc）：A collection of installable packages and libraries used for distributing and working with PostgresML. 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`packages/README.md` - **PostgresML Dashboard**（project_doc）：PostgresML provides a dashboard with analytical views of the training data and model performance, as well as integrated notebooks for rapid iteration. It is primarily written in Rust using Rocket https://rocket.rs/ as a lightweight web framework and SQLx https://github.com/launchbadge/sqlx to interact with the database. 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`pgml-dashboard/README.md` - **Readme**（project_doc）：Please see the quick start instructions https://postgresml.org/docs/open-source/pgml/developers/quick-start-with-docker for general information on installing or deploying PostgresML. A developer guide https://postgresml.org/docs/open-source/pgml/developers/contributing is also available for those who would like to contribute. 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`pgml-extension/README.md` - **pgml-components**（project_doc）：pgml-components is a CLI for working with Rust web apps written with Rocket, Sailfish and SQLx, our toolkit of choice. It's currently a work in progress and only used internally by us, but the long term goal is to make it into a comprehensive framework for building web apps in Rust. 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`packages/cargo-pgml-components/README.md` - **pgml-rds-proxy**（project_doc）：A pgcat-based PostgreSQL proxy that allows to use PostgresML functions on managed PostgreSQL databases that may not have Internet access, like AWS RDS. 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`packages/pgml-rds-proxy/README.md` - **Terraform configuration for pgml-rds-proxy on EC2**（project_doc）：Terraform configuration for pgml-rds-proxy on EC2 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`packages/pgml-rds-proxy/ec2/README.md` - **PostgresML Chatbot Builder**（project_doc）：PostgresML Chatbot Builder A command line tool to build and deploy a knowledge based chatbot using PostgresML and OpenAI API. 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`pgml-apps/pgml-chat/README.md` - **Home**（project_doc）：recent blog posts 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`pgml-cms/blog/README.md` - **Careers**（project_doc）：PostgresML is building a GPU-powered AI application database. You can perform microsecond inference with the world's most capable feature store. It allows you to easily train and deploy online models using only SQL. We're looking for an experienced Engineers to help shape the core product, inside and out. We're generally looking for individual contributors, but everyone be critical in building the future team as wel… 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`pgml-cms/careers/README.md` - **Python vs. PostgresML**（project_doc）：postgresql \copy flights delay mat FROM '~/Desktop/flights.csv' CSV HEADER; 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`pgml-dashboard/content/blog/benchmarks/python_microservices_vs_postgresml/README.md` - **Bootstrap 5**（project_doc）：Sleek, intuitive, and powerful front-end framework for faster and easier web development. Explore Bootstrap docs » Report bug · Request feature · Themes · Blog 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`pgml-dashboard/static/css/bootstrap-5.3.0-alpha1/README.md` - **Prerequisites**（project_doc）：In this tutorial, we will show how to build a text processing pipeline using PostgresML and dbt data build tool . We will use PostgresML to chunk documents and compute embeddings and dbt to execute different steps in a specific order figure below . 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`pgml-extension/examples/dbt/embeddings/README.md` - **Open Source Alternative for Building End-to-End Vector Search Applications without OpenAI & Pinecone**（project_doc）：Open Source Alternative for Building End-to-End Vector Search Applications without OpenAI & Pinecone 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`pgml-sdks/pgml/README.md` - **Open Source Alternative for Building End-to-End Vector Search Applications without OpenAI & Pinecone**（project_doc）：Open Source Alternative for Building End-to-End Vector Search Applications without OpenAI & Pinecone 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`pgml-sdks/pgml/javascript/README.md` - **Examples**（project_doc）：Prerequisites Before running any examples first install dependencies and set the DATABASE URL environment variable: 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`pgml-sdks/pgml/javascript/examples/README.md` - **Webpack Demo**（project_doc）：The JavaScript SDK utilizes native node modules as our SDK is written in Rust. To get it working with webpack, we need a loader that is designed to work with native node modules. In this case, we have opted to use the node-loader https://github.com/webpack-contrib/node-loader module. See webpack.config.js ./webpack.config.js for how we configured it. 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`pgml-sdks/pgml/javascript/examples/webpack/README.md` - **Open Source Alternative for Building End-to-End Vector Search Applications without OpenAI & Pinecone**（project_doc）：Open Source Alternative for Building End-to-End Vector Search Applications without OpenAI & Pinecone 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`pgml-sdks/pgml/python/README.md` - **Examples**（project_doc）：Prerequisites Before running any examples first install dependencies and set the DATABASE URL environment variable: 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`pgml-sdks/pgml/python/examples/README.md` - **A Tool for Automatically Translating to Py03 and Neon compatible Rust**（project_doc）：A Tool for Automatically Translating to Py03 and Neon compatible Rust 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`pgml-sdks/rust-bridge/README.md` - **Coming Soon**（project_doc）： 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`pgml-sdks/rust-bridge/examples/README.md` - **Contributing**（project_doc）：Thank you for your interest in contributing to PostgresML! We are an open source, MIT licensed project, and we welcome all contributions, including bug fixes, features, documentation, typo fixes, and Github stars. 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`pgml-cms/docs/open-source/pgml/developers/contributing.md` - **Table of contents**（project_doc）：Overview README.md Getting started introduction/getting-started/README.md Create your database introduction/getting-started/create-your-database.md Connect your app introduction/getting-started/connect-your-app.md Import your data introduction/import-your-data/README.md Logical replication introduction/import-your-data/logical-replication/README.md Foreign Data Wrappers introduction/import-your-data/foreign-data-wra… 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`pgml-cms/docs/SUMMARY.md` - **Why PostgresML?**（project_doc）：PostgresML offers a unique and modern architecture which replaces service-based machine learning applications with a single database. The benefits of this approach are measurable in performance, ease of use, and data integrity. 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`pgml-cms/docs/TODO/architecture/why-postgresml.md` - **Example Application**（project_doc）：CLI tool to build and deploy chatbots 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`pgml-cms/docs/TODO/chatbots.md` - **GGML Quantized LLM support for Huggingface Transformers**（project_doc）：Quantization allows PostgresML to fit larger models in less RAM. 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`pgml-cms/docs/TODO/ggml-quantized-llm-support-for-huggingface-transformers.md` - **Dedicated**（project_doc）：Dedicated databases are created on dedicated hardware in our hosting provider currently AWS EC2 and have guaranteed capacity and basically limitless horizontal scalability. PostgresML supports up to 16 Postgres replicas, 16 PgCat poolers and petabytes of disk storage, allowing teams that use it to scale to millions of requests per second at a click of a button. 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`pgml-cms/docs/cloud/dedicated.md` - **Teams**（project_doc）：Invite additional team members to manage your databases. 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`pgml-cms/docs/cloud/enterprise/teams.md` - **VPC**（project_doc）：PostgresML can be launched in your Virtual Private Cloud VPC account on AWS, Azure or GCP. 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`pgml-cms/docs/cloud/enterprise/vpc.md` - **PostgresML Cloud**（project_doc）：PostgresML Cloud is the best place to perform in-database ML/AI. 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`pgml-cms/docs/cloud/overview.md` - **Privacy Policy**（project_doc）：This privacy policy “Policy” describes how Hyperparam Inc. “Company”, “PostgresML”, “we”, “us” collects, uses, and shares personal information of consumer users of this website, https://postgresml.org the “Site” , as well as associated products and services together, the “Services” , and applies to personal information that we collect through the Site and our Services as well as personal information you provide to u… 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`pgml-cms/docs/cloud/privacy-policy.md` - **Serverless**（project_doc）：A Serverless PostgresML database can be created in less than 5 seconds and provides immediate access to modern GPU acceleration, a predefined set of state-of-the-art large language models that should satisfy most use cases, and dozens of supervised learning algorithms like XGBoost, LightGBM, Catboost, and everything from Scikit-learn. We call this combination of tools an AI engine. With a Serverless engine, storage… 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`pgml-cms/docs/cloud/serverless.md` - **Terms of Service**（project_doc）：Welcome to PostgresML! Your use of PostgresML’s services, including the services PostgresML makes available through this website and applications which link to these terms of service the “Site” and to all software or services offered by PostgresML in connection with any of those the “Services” , is governed by these terms of service the “Terms” , so please carefully read them before using the Services. For the purpo… 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`pgml-cms/docs/cloud/terms-of-service.md` - **FAQ**（project_doc）：PostgresML Frequently Asked Questions 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`pgml-cms/docs/introduction/faq.md` - **Connect your app**（project_doc）：Connect your application to PostgresML using our SDK or any standard PostgreSQL client. 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`pgml-cms/docs/introduction/getting-started/connect-your-app.md` - **Create your database**（project_doc）：- 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`pgml-cms/docs/introduction/getting-started/create-your-database.md` - **Move data with COPY**（project_doc）：Move data into PostgresML from data files using COPY and CSV. 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`pgml-cms/docs/introduction/import-your-data/copy.md` - **Foreign Data Wrappers**（project_doc）：Connect your production database to PostgresML using Foreign Data Wrappers. 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`pgml-cms/docs/introduction/import-your-data/foreign-data-wrappers.md` - **Connect your VPC to PostgresML**（project_doc）：If your database doesn't have Internet access, PostgresML will need a service to proxy connections to your database. Any TCP proxy will do, and we also provide an nginx-based Docker image than can be used without any additional configuration. 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`pgml-cms/docs/introduction/import-your-data/logical-replication/inside-a-vpc.md` - **Migrate with pg dump**（project_doc）：Migrate your PostgreSQL database to PostgresML using pg dump. 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`pgml-cms/docs/introduction/import-your-data/pg-dump.md` - **Documents**（project_doc）：Documents are a type of data that has no predefined schema. Typically stored as JSON, documents can be used to allow users of your application to store, retrieve and work with arbitrary data. 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`pgml-cms/docs/introduction/import-your-data/storage-and-retrieval/documents.md` - **LLM based pipelines with PostgresML and dbt data build tool**（project_doc）：LLM based pipelines with PostgresML and dbt data build tool 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`pgml-cms/docs/introduction/import-your-data/storage-and-retrieval/llm-based-pipelines-with-postgresml-and-dbt-data-build-tool.md` - **Partitioning**（project_doc）：Partitioning is splitting Posgres tables into multiple smaller tables, with the intention of querying each smaller table independently. This is useful and sometimes necessary when tables get so large that accessing them becomes too slow. Partitioning requires detailed knowledge of the dataset and uses that knowledge to help Postgres execute queries faster. 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`pgml-cms/docs/introduction/import-your-data/storage-and-retrieval/partitioning.md` - **Tabular Data**（project_doc）： 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`pgml-cms/docs/introduction/import-your-data/storage-and-retrieval/tabular-data.md` - **Collections**（project_doc）：- 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`pgml-cms/docs/open-source/korvus/api/collections.md` - **Pipelines**（project_doc）：- 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`pgml-cms/docs/open-source/korvus/api/pipelines.md` - **Rag with OpenAI**（project_doc）：An example application performing RAG with Korvus and OpenAI. 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`pgml-cms/docs/open-source/korvus/example-apps/rag-with-openai.md` - **Semantic Search**（project_doc）：- 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`pgml-cms/docs/open-source/korvus/example-apps/semantic-search.md` - **Constructing Pipelines**（project_doc）：Pipelines are a powerful feature for processing and preparing documents for efficient search and retrieval. They define a series of transformations applied to your data, enabling operations like text splitting, semantic embedding, and full-text search preparation. This guide will walk you through the process of constructing Pipeline schemas, allowing you to customize how your documents are processed and indexed. 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`pgml-cms/docs/open-source/korvus/guides/constructing-pipelines.md` - **Document Search**（project_doc）：Korvus is specifically designed to provide powerful, flexible document search. Pipeline s are required to perform search. See the Pipelines docs/api/client-sdk/pipelines for more information about using Pipeline s. 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`pgml-cms/docs/open-source/korvus/guides/document-search.md` - **OpenSourceAI**（project_doc）：OpenSourceAI is a drop in replacement for OpenAI's chat completion endpoint. 激活提示：当用户需要理解项目结构、安装方式或边界时参考。 证据：`pgml-cms/docs/open-source/korvus/guides/opensourceai.md` ## 证据索引 - 共索引 79 条证据。 - **Overview**（documentation）：PostgresML is a complete MLOps platform built inside PostgreSQL. Our operating principle is: 证据：`pgml-cms/docs/README.md` - **PostgresML architecture**（documentation）：PostgresML is an extension for the PostgreSQL database server. It operates inside the database, using the same hardware to perform machine learning tasks. 证据：`pgml-cms/docs/TODO/architecture/README.md` - **Enterprise**（documentation）：Enterprise plans are ideal large companies that have special compliance needs and deployment configurations; with options for cloud-prem VPC , on-prem, ACL’s and more. 证据：`pgml-cms/docs/cloud/enterprise/README.md` - **Getting started**（documentation）：A PostgresML deployment consists of multiple components working in concert to provide a complete Machine Learning platform: 证据：`pgml-cms/docs/introduction/getting-started/README.md` - **Import your data**（documentation）：AI needs data, whether it's generating text with LLMs, creating embeddings, or training regression or classification models on customer data. 证据：`pgml-cms/docs/introduction/import-your-data/README.md` - **Logical replication**（documentation）：Logical replication allows your PostgresML database to copy data from your primary database to PostgresML in real time. As soon as your customers make changes to their data on your website, those changes will become available in PostgresML. 证据：`pgml-cms/docs/introduction/import-your-data/logical-replication/README.md` - **Tabular data**（documentation）：Tabular data is data stored in tables. A table is a format that defines rows and columns, and is the most common type of data organization. Examples of tabular data are spreadsheets, database tables, CSV files, and Pandas dataframes. 证据：`pgml-cms/docs/introduction/import-your-data/storage-and-retrieval/README.md` - **Korvus**（documentation）：Korvus is an all-in-one, open-source RAG Retrieval-Augmented Generation pipeline built for PostgresML. It combines LLMs, vector memory, embedding generation, reranking, summarization and custom models into a single query, maximizing performance and simplifying your search architecture. 证据：`pgml-cms/docs/open-source/korvus/README.md` - **API**（documentation）：The API docs provide a brief overview of the available methods for Korvus Classes / Structs. 证据：`pgml-cms/docs/open-source/korvus/api/README.md` - **Example Applications**（documentation）：These example apps cover some common use cases. 证据：`pgml-cms/docs/open-source/korvus/example-apps/README.md` - **Guides**（documentation）：These guides cover some more complex examples for using the available methods in Korvus. 证据：`pgml-cms/docs/open-source/korvus/guides/README.md` - **PgCat pooler**（documentation）：PgCat is PostgreSQL connection pooler and proxy which scales PostgreSQL and PostgresML databases beyond a single instance. It supports replicas, load balancing, sharding, failover, and many more features expected out of high availability enterprise-grade PostgreSQL deployment. Written in Rust using Tokio, it takes advantage of multiple CPUs and the safety and performance guarantees of the Rust language. 证据：`pgml-cms/docs/open-source/pgcat/README.md` - **SQL extension**（documentation）：pgml is a PostgreSQL extension which adds SQL functions to the database. Those functions provide access to AI models downloaded from Hugging Face, and classical machine learning algorithms like XGBoost and LightGBM. 证据：`pgml-cms/docs/open-source/pgml/README.md` - **PGML API**（documentation）：The API docs provides a brief overview of the available functions exposed by pgml . 证据：`pgml-cms/docs/open-source/pgml/api/README.md` - **pgml.predict**（documentation）：The pgml.predict function is the key value proposition of PostgresML. It provides online predictions using the best, automatically deployed model for a project. The API for predictions is very simple and only requires two arguments: the project name and the features used for prediction. 证据：`pgml-cms/docs/open-source/pgml/api/pgml.predict/README.md` - **Developers**（documentation）：Documentation relevant to self-hosting, compiling or contributing to PostgresML 证据：`pgml-cms/docs/open-source/pgml/developers/README.md` - **Self-hosting**（documentation）：PostgresML is a Postgres extension, so running it is very similar to running a self-hosted PostgreSQL database server. A typical architecture consists of a primary database that will serve reads and writes, optional replicas to scale reads horizontally, and a pooler to load balance connections. 证据：`pgml-cms/docs/open-source/pgml/developers/self-hosting/README.md` - **Guides**（documentation）：Long form examples demonstrating use cases for PostgresML 证据：`pgml-cms/docs/open-source/pgml/guides/README.md` - **Chatbots**（documentation）：This tutorial seeks to broadly cover the majority of topics required to not only implement a modern chatbot, but understand why we build them this way. There are three primary sections: 证据：`pgml-cms/docs/open-source/pgml/guides/chatbots/README.md` - **Embeddings**（documentation）：As the demand for sophisticated data analysis and machine learning capabilities within databases grows, so does the need for efficient and scalable solutions. PostgresML offers a powerful platform for integrating machine learning directly into PostgreSQL, enabling you to perform complex computations and predictive analytics without ever leaving your database. 证据：`pgml-cms/docs/open-source/pgml/guides/embeddings/README.md` - **LLMs**（documentation）：PostgresML integrates 🤗 Hugging Face Transformers https://huggingface.co/transformers to bring state-of-the-art models into the data layer. There are tens of thousands of pre-trained models with pipelines to turn raw inputs into useful results. Many state of the art deep learning architectures have been published and made available for download. You will want to browse all the models https://huggingface.co/models available to find the perfect solution for your dataset https://huggingface.co/dataset and task https://huggingface.co/tasks . For instance, with PostgresML you can: 证据：`pgml-cms/docs/open-source/pgml/guides/llms/README.md` - **Supervised Learning**（documentation）：A large part of the machine learning workflow is acquiring, cleaning, and preparing data for training algorithms. Naturally, we think Postgres is a great place to store your data. For the purpose of this example, we'll load a toy dataset, the classic handwritten digits image collection, from scikit-learn. 证据：`pgml-cms/docs/open-source/pgml/guides/supervised-learning/README.md` - **Architecture**（documentation）：Postgres + GPUs for ML/AI applications. 证据：`README.md` - **Packages**（documentation）：A collection of installable packages and libraries used for distributing and working with PostgresML. 证据：`packages/README.md` - **PostgresML Dashboard**（documentation）：PostgresML provides a dashboard with analytical views of the training data and model performance, as well as integrated notebooks for rapid iteration. It is primarily written in Rust using Rocket https://rocket.rs/ as a lightweight web framework and SQLx https://github.com/launchbadge/sqlx to interact with the database. 证据：`pgml-dashboard/README.md` - **Readme**（documentation）：Please see the quick start instructions https://postgresml.org/docs/open-source/pgml/developers/quick-start-with-docker for general information on installing or deploying PostgresML. A developer guide https://postgresml.org/docs/open-source/pgml/developers/contributing is also available for those who would like to contribute. 证据：`pgml-extension/README.md` - **pgml-components**（documentation）：pgml-components is a CLI for working with Rust web apps written with Rocket, Sailfish and SQLx, our toolkit of choice. It's currently a work in progress and only used internally by us, but the long term goal is to make it into a comprehensive framework for building web apps in Rust. 证据：`packages/cargo-pgml-components/README.md` - **pgml-rds-proxy**（documentation）：A pgcat-based PostgreSQL proxy that allows to use PostgresML functions on managed PostgreSQL databases that may not have Internet access, like AWS RDS. 证据：`packages/pgml-rds-proxy/README.md` - **Terraform configuration for pgml-rds-proxy on EC2**（documentation）：Terraform configuration for pgml-rds-proxy on EC2 证据：`packages/pgml-rds-proxy/ec2/README.md` - **PostgresML Chatbot Builder**（documentation）：PostgresML Chatbot Builder A command line tool to build and deploy a knowledge based chatbot using PostgresML and OpenAI API. 证据：`pgml-apps/pgml-chat/README.md` - **Home**（documentation）：What's Hacker News' problem with open source AI whats-hacker-news-problem-with-open-source-ai.md "mention" announcing-support-for-meta-llama-3.1 announcing-support-for-meta-llama-3.1.md "mention" announcing-the-release-of-our-rust-sdk announcing-the-release-of-our-rust-sdk.md "mention" meet-us-at-the-2024-ai-dev-summit-conference meet-us-at-the-2024-ai-dev-summit-conference.md "mention" introducing-the-openai-switch-kit-move-from-closed-to-open-source-ai-in-minutes.md introducing-the-openai-switch-kit-move-from-closed-to-open-source-ai-in-minutes.md "mention" speeding-up-vector-recall-5x-with-hnsw.md speeding-up-vector-recall-5x-with-hnsw.md "mention" how-to-improve-search-results-with-mach… 证据：`pgml-cms/blog/README.md` - **Careers**（documentation）：PostgresML is building a GPU-powered AI application database. You can perform microsecond inference with the world's most capable feature store. It allows you to easily train and deploy online models using only SQL. We're looking for an experienced Engineers to help shape the core product, inside and out. We're generally looking for individual contributors, but everyone be critical in building the future team as well as the core product, while leading efforts toward more efficient and effective Machine Learning workflows for our customers. 证据：`pgml-cms/careers/README.md` - **Python vs. PostgresML**（documentation）：postgresql \copy flights delay mat FROM '~/Desktop/flights.csv' CSV HEADER; 证据：`pgml-dashboard/content/blog/benchmarks/python_microservices_vs_postgresml/README.md` - **Bootstrap 5**（documentation）：Sleek, intuitive, and powerful front-end framework for faster and easier web development. Explore Bootstrap docs » Report bug · Request feature · Themes · Blog 证据：`pgml-dashboard/static/css/bootstrap-5.3.0-alpha1/README.md` - **Prerequisites**（documentation）：In this tutorial, we will show how to build a text processing pipeline using PostgresML and dbt data build tool . We will use PostgresML to chunk documents and compute embeddings and dbt to execute different steps in a specific order figure below . 证据：`pgml-extension/examples/dbt/embeddings/README.md` - **Open Source Alternative for Building End-to-End Vector Search Applications without OpenAI & Pinecone**（documentation）：Open Source Alternative for Building End-to-End Vector Search Applications without OpenAI & Pinecone 证据：`pgml-sdks/pgml/README.md` - **Open Source Alternative for Building End-to-End Vector Search Applications without OpenAI & Pinecone**（documentation）：Open Source Alternative for Building End-to-End Vector Search Applications without OpenAI & Pinecone 证据：`pgml-sdks/pgml/javascript/README.md` - **Examples**（documentation）：Prerequisites Before running any examples first install dependencies and set the DATABASE URL environment variable: 证据：`pgml-sdks/pgml/javascript/examples/README.md` - **Webpack Demo**（documentation）：The JavaScript SDK utilizes native node modules as our SDK is written in Rust. To get it working with webpack, we need a loader that is designed to work with native node modules. In this case, we have opted to use the node-loader https://github.com/webpack-contrib/node-loader module. See webpack.config.js ./webpack.config.js for how we configured it. 证据：`pgml-sdks/pgml/javascript/examples/webpack/README.md` - **Open Source Alternative for Building End-to-End Vector Search Applications without OpenAI & Pinecone**（documentation）：Open Source Alternative for Building End-to-End Vector Search Applications without OpenAI & Pinecone 证据：`pgml-sdks/pgml/python/README.md` - **Examples**（documentation）：Prerequisites Before running any examples first install dependencies and set the DATABASE URL environment variable: 证据：`pgml-sdks/pgml/python/examples/README.md` - **A Tool for Automatically Translating to Py03 and Neon compatible Rust**（documentation）：A Tool for Automatically Translating to Py03 and Neon compatible Rust 证据：`pgml-sdks/rust-bridge/README.md` - **Coming Soon**（documentation）：Coming Soon 证据：`pgml-sdks/rust-bridge/examples/README.md` - **Contributing**（documentation）：Thank you for your interest in contributing to PostgresML! We are an open source, MIT licensed project, and we welcome all contributions, including bug fixes, features, documentation, typo fixes, and Github stars. 证据：`pgml-cms/docs/open-source/pgml/developers/contributing.md` - **Package**（package_manifest）：{ "dependencies": { "@codemirror/lang-javascript": "^6.2.1", "@codemirror/lang-python": "^6.1.3", "@codemirror/lang-rust": "^6.0.1", "@codemirror/lang-cpp": "^6.0.2", "postgresml-lang-sql": "^6.6.3-5", "@codemirror/lang-json": "^6.0.1", "@codemirror/state": "^6.2.1", "@codemirror/view": "^6.21.0", "codemirror": "^6.0.1", "autosize": "^6.0.1", "dompurify": "^3.0.6", "marked": "^9.1.0" } } 证据：`pgml-dashboard/package.json` - **Package**（package_manifest）：{ "name": "pgml-cli", "version": "0.10.0", "description": "CLI for PostgresML, the GPU-powered AI application database.", "keywords": "postgres", "machine learning", "vector databases", "embeddings" , "bin": { "pgml": "index.js" }, "author": { "name": "PostgresML", "email": "team@postgresml.org", "url": "https://postgresml.org" }, "repository": { "type": "git", "url": "https://github.com/postgresml/postgresml" }, "license": "MIT", "dependencies": { "pgml": "0.10.0" } } 证据：`pgml-sdks/pgml/javascript-cli/package.json` - **Package**（package_manifest）：{ "name": "getting-started", "version": "1.0.0", "description": "", "main": "index.js", "scripts": { "test": "echo \"Error: no test specified\" && exit 1" }, "author": "", "license": "ISC", "dependencies": { "dotenv": "^16.3.1", "pgml": "^1.0.0" } } 证据：`pgml-sdks/pgml/javascript/examples/package.json` - **Package**（package_manifest）：{ "name": "webpack", "version": "1.0.0", "description": "", "main": "index.js", "scripts": { "test": "echo \"Error: no test specified\" && exit 1", "build": "webpack" }, "author": "", "license": "ISC", "devDependencies": { "node-loader": "^2.0.0", "webpack": "^5.88.2", "webpack-cli": "^5.1.4" }, "dependencies": { "dotenv": "^16.3.1", "pgml": "^1.0.0" } } 证据：`pgml-sdks/pgml/javascript/examples/webpack/package.json` - **Package**（package_manifest）：{ "name": "pgml", "version": "1.1.1", "description": "Open Source Alternative for Building End-to-End Vector Search Applications without OpenAI & Pinecone", "keywords": "postgres", "machine learning", "vector databases", "embeddings" , "main": "index.js", "scripts": { "build": "node build.js", "build-release": "node build.js --release" }, "author": { "name": "PostgresML", "email": "team@postgresml.org", "url": "https://postgresml.org" }, "repository": { "type": "git", "url": "https://github.com/postgresml/postgresml" }, "license": "MIT", "devDependencies": { "@types/node": "^20.3.1", "cargo-cp-artifact": "^0.1" }, "dependencies": { "dotenv": "^16.4.4" } } 证据：`pgml-sdks/pgml/javascript/package.json` - **Package**（package_manifest）：{ "name": "pgml-tests", "version": "0.1.0", "description": "", "type": "module", "scripts": { "test": "NODE OPTIONS=--experimental-vm-modules jest" }, "devDependencies": { "@types/jest": "^29.5.3", "jest": "^29.6.1", "ts-jest": "^29.1.1", "typescript": "^5.1.6" } } 证据：`pgml-sdks/pgml/javascript/tests/package.json` - **License**（source_file）：Copyright c 2011-2022 The Bootstrap Authors 证据：`pgml-dashboard/static/css/bootstrap-5.3.0-alpha1/LICENSE` - **Table of contents**（documentation）：Overview README.md Getting started introduction/getting-started/README.md Create your database introduction/getting-started/create-your-database.md Connect your app introduction/getting-started/connect-your-app.md Import your data introduction/import-your-data/README.md Logical replication introduction/import-your-data/logical-replication/README.md Foreign Data Wrappers introduction/import-your-data/foreign-data-wrappers.md Move data with COPY introduction/import-your-data/copy.md Migrate with pg dump introduction/import-your-data/pg-dump.md Storage & Retrieval introduction/import-your-data/storage-and-retrieval/README.md Documents introduction/import-your-data/storage-and-retrieval/documen… 证据：`pgml-cms/docs/SUMMARY.md` - **Why PostgresML?**（documentation）：PostgresML offers a unique and modern architecture which replaces service-based machine learning applications with a single database. The benefits of this approach are measurable in performance, ease of use, and data integrity. 证据：`pgml-cms/docs/TODO/architecture/why-postgresml.md` - **Example Application**（documentation）：A command line tool to build and deploy a knowledge based chatbot using PostgresML and OpenAI API. 证据：`pgml-cms/docs/TODO/chatbots.md` - **GGML Quantized LLM support for Huggingface Transformers**（documentation）：GGML Quantized LLM support for Huggingface Transformers 证据：`pgml-cms/docs/TODO/ggml-quantized-llm-support-for-huggingface-transformers.md` - **Dedicated**（documentation）：Dedicated databases are created on dedicated hardware in our hosting provider currently AWS EC2 and have guaranteed capacity and basically limitless horizontal scalability. PostgresML supports up to 16 Postgres replicas, 16 PgCat poolers and petabytes of disk storage, allowing teams that use it to scale to millions of requests per second at a click of a button. 证据：`pgml-cms/docs/cloud/dedicated.md` - **Teams**（documentation）：Invite additional team members to manage your databases. 证据：`pgml-cms/docs/cloud/enterprise/teams.md` - **VPC**（documentation）：PostgresML can be launched in your Virtual Private Cloud VPC account on AWS, Azure or GCP. 证据：`pgml-cms/docs/cloud/enterprise/vpc.md` - **PostgresML Cloud**（documentation）：PostgresML Cloud is the best place to perform in-database ML/AI. 证据：`pgml-cms/docs/cloud/overview.md` - **Privacy Policy**（documentation）：This privacy policy “Policy” describes how Hyperparam Inc. “Company”, “PostgresML”, “we”, “us” collects, uses, and shares personal information of consumer users of this website, https://postgresml.org the “Site” , as well as associated products and services together, the “Services” , and applies to personal information that we collect through the Site and our Services as well as personal information you provide to us directly. This Policy also applies to any of our other websites that post this Policy. Please note that by using the Site or the Services, you accept the practices and policies described in this Policy and you consent that we will collect, use, and share your personal informati… 证据：`pgml-cms/docs/cloud/privacy-policy.md` - 其余 19 条证据见 `AI_CONTEXT_PACK.json` 或 `EVIDENCE_INDEX.json`。 ## 宿主 AI 必须遵守的规则 - **把本资产当作开工前上下文，而不是运行环境。**：AI Context Pack 只包含证据化项目理解，不包含目标项目的可执行状态。 证据：`pgml-cms/docs/README.md`, `pgml-cms/docs/TODO/architecture/README.md`, `pgml-cms/docs/cloud/enterprise/README.md` - **回答用户时区分可预览内容与必须安装后才能验证的内容。**：安装前体验的消费者价值来自降低误装和误判，而不是伪装成真实运行。 证据：`pgml-cms/docs/README.md`, `pgml-cms/docs/TODO/architecture/README.md`, `pgml-cms/docs/cloud/enterprise/README.md` ## 用户开工前应该回答的问题 - 你准备在哪个宿主 AI 或本地环境中使用它？ - 你只是想先体验工作流，还是准备真实安装？ - 你最在意的是安装成本、输出质量、还是和现有规则的冲突？ ## 验收标准 - 所有能力声明都能回指到 evidence_refs 中的文件路径。 - AI_CONTEXT_PACK.md 没有把预览包装成真实运行。 - 用户能在 3 分钟内看懂适合谁、能做什么、如何开始和风险边界。 --- ## Doramagic Context Augmentation 下面内容用于强化 Repomix/AI Context Pack 主体。Human Manual 只提供阅读骨架；踩坑日志会被转成宿主 AI 必须遵守的工作约束。 ## Human Manual 骨架 使用规则：这里只是项目阅读路线和显著性信号，不是事实权威。具体事实仍必须回到 repo evidence / Claim Graph。 宿主 AI 硬性规则： - 不得把页标题、章节顺序、摘要或 importance 当作项目事实证据。 - 解释 Human Manual 骨架时，必须明确说它只是阅读路线/显著性信号。 - 能力、安装、兼容性、运行状态和风险判断必须引用 repo evidence、source path 或 Claim Graph。 - **Overview and System Architecture**：importance `high` - source_paths: README.md, docker/Dockerfile, docker/entrypoint.sh, pgml-extension/Cargo.toml, pgml-extension/build.rs - **Core Extension: pgml-extension Internals**：importance `high` - source_paths: pgml-extension/src/lib.rs, pgml-extension/src/api.rs, pgml-extension/src/config.rs, pgml-extension/src/metrics.rs, pgml-extension/src/vectors.rs - **RAG Pipeline, Transformers & LLM Integration**：importance `high` - source_paths: pgml-extension/src/bindings/transformers/mod.rs, pgml-extension/src/bindings/transformers/transform.rs, pgml-extension/src/bindings/transformers/whitelist.rs, pgml-extension/src/bindings/transformers/transformers.py, pgml-extension/src/bindings/langchain/mod.rs - **Deployment, SDKs & Operations**：importance `high` - source_paths: docker/Dockerfile, docker/entrypoint.sh, docker/dashboard.sh, docker/local_dev.conf, docker/pg_hba.conf ## Repo Inspection Evidence / 源码检查证据 - repo_clone_verified: true - repo_inspection_verified: true - repo_commit: `caf2b6ccdf0d6efc2c1910cbc06725a34320181a` - inspected_files: `README.md`, `packages/README.md`, `packages/cargo-pgml-components/Cargo.toml`, `packages/cargo-pgml-components/README.md`, `packages/cargo-pgml-components/sailfish.toml`, `packages/pgml-components/Cargo.toml`, `packages/pgml-rds-proxy/README.md`, `packages/pgml-rds-proxy/ec2/README.md` 宿主 AI 硬性规则： - 没有 repo_clone_verified=true 时，不得声称已经读过源码。 - 没有 repo_inspection_verified=true 时，不得把 README/docs/package 文件判断写成事实。 - 没有 quick_start_verified=true 时，不得声称 Quick Start 已跑通。 ## Doramagic Pitfall Constraints / 踩坑约束 这些规则来自 Doramagic 发现、验证或编译过程中的项目专属坑点。宿主 AI 必须把它们当作工作约束，而不是普通说明文字。 ### Constraint 1: 来源证据：Invalid pgml.so binary for version 2.10 amd64 .deb package - Trigger: GitHub 社区证据显示该项目存在一个安装相关的待验证问题：Invalid pgml.so binary for version 2.10 amd64 .deb package - Why it matters: 可能增加新用户试用和生产接入成本。 - Evidence: community_evidence:github | https://github.com/postgresml/postgresml/issues/1674 | 来源类型 github_issue 暴露的待验证使用条件。 - Hard boundary: 不要把这个坑点包装成已解决、已验证或可忽略，除非后续验证证据明确证明它已经关闭。 ### Constraint 2: 来源证据：Issues with Docker container on port 5432 - Trigger: GitHub 社区证据显示该项目存在一个安装相关的待验证问题：Issues with Docker container on port 5432 - Why it matters: 可能增加新用户试用和生产接入成本。 - Evidence: community_evidence:github | https://github.com/postgresml/postgresml/issues/1392 | 来源讨论提到 docker 相关条件，需在安装/试用前复核。 - Hard boundary: 不要把这个坑点包装成已解决、已验证或可忽略，除非后续验证证据明确证明它已经关闭。 ### Constraint 3: 来源证据：Performing embedding inference as part of an UPDATE crashes the server within the docker container with an "illegal ins… - Trigger: GitHub 社区证据显示该项目存在一个安装相关的待验证问题：Performing embedding inference as part of an UPDATE crashes the server within the docker container with an "illegal instruction" - Why it matters: 可能阻塞安装或首次运行。 - Evidence: community_evidence:github | https://github.com/postgresml/postgresml/issues/1515 | 来源讨论提到 docker 相关条件，需在安装/试用前复核。 - Hard boundary: 不要把这个坑点包装成已解决、已验证或可忽略，除非后续验证证据明确证明它已经关闭。 ### Constraint 4: 来源证据：Working docker image for latest release for ubuntu 24 - Trigger: GitHub 社区证据显示该项目存在一个安装相关的待验证问题：Working docker image for latest release for ubuntu 24 - Why it matters: 可能增加新用户试用和生产接入成本。 - Evidence: community_evidence:github | https://github.com/postgresml/postgresml/issues/1680 | 来源讨论提到 python 相关条件，需在安装/试用前复核。 - Hard boundary: 不要把这个坑点包装成已解决、已验证或可忽略，除非后续验证证据明确证明它已经关闭。 ### Constraint 5: 来源证据：Invalid model can be "deployed", if there is no prior trained model for a project - Trigger: GitHub 社区证据显示该项目存在一个配置相关的待验证问题：Invalid model can be "deployed", if there is no prior trained model for a project - Why it matters: 可能增加新用户试用和生产接入成本。 - Evidence: community_evidence:github | https://github.com/postgresml/postgresml/issues/628 | 来源类型 github_issue 暴露的待验证使用条件。 - Hard boundary: 不要把这个坑点包装成已解决、已验证或可忽略，除非后续验证证据明确证明它已经关闭。 ### Constraint 6: 来源证据：Add model Export & Load as part of the full model lifecycle - Trigger: GitHub 社区证据显示该项目存在一个运行相关的待验证问题：Add model Export & Load as part of the full model lifecycle - Why it matters: 可能增加新用户试用和生产接入成本。 - Evidence: community_evidence:github | https://github.com/postgresml/postgresml/issues/1686 | 来源类型 github_issue 暴露的待验证使用条件。 - Hard boundary: 不要把这个坑点包装成已解决、已验证或可忽略，除非后续验证证据明确证明它已经关闭。 ### Constraint 7: 安装或运行可能执行高权限 shell - Trigger: 项目文本出现 postinstall/sudo/curl pipe/shell 等关键词。 - Host AI rule: 扫描 package scripts、安装脚本和 README 命令。 - Why it matters: 用户本机文件、凭证或环境可能受影响，需要先隔离验证。 - Evidence: packet_text.keyword_scan | https://github.com/postgresml/postgresml | matched postinstall / sudo / curl pipe / shell keyword - Hard boundary: 不要把这个坑点包装成已解决、已验证或可忽略，除非后续验证证据明确证明它已经关闭。 ### Constraint 8: 依赖 Docker 环境 - Trigger: 安装/运行入口包含 Docker 命令：docker run -it -v postgresml_data:/var/lib/postgresql -p 5433:5432 -p 8000:8000 ghcr.io/postgresml/postgresml:2.10.0 sudo -u postgresml psql -d postgresml - Host AI rule: 标注 Docker 前置条件，并提供非 Docker 路径或失败提示。 - Why it matters: 非工程用户可能没有 Docker，启动成本明显增加。 - Evidence: identity.distribution | https://github.com/postgresml/postgresml | docker run -it -v postgresml_data:/var/lib/postgresql -p 5433:5432 -p 8000:8000 ghcr.io/postgresml/postgresml:2.10.0 sudo -u postgresml psql -d postgresml - Hard boundary: 不要把这个坑点包装成已解决、已验证或可忽略，除非后续验证证据明确证明它已经关闭。 ### Constraint 9: 来源证据：Best way to install postgresML on windows - Trigger: GitHub 社区证据显示该项目存在一个安装相关的待验证问题：Best way to install postgresML on windows - Why it matters: 可能增加新用户试用和生产接入成本。 - Evidence: community_evidence:github | https://github.com/postgresml/postgresml/issues/1697 | 来源讨论提到 docker 相关条件，需在安装/试用前复核。 - Hard boundary: 不要把这个坑点包装成已解决、已验证或可忽略，除非后续验证证据明确证明它已经关闭。 ### Constraint 10: 来源证据：OSError: We couldn't connect to 'https://huggingface.co' - Trigger: GitHub 社区证据显示该项目存在一个安装相关的待验证问题：OSError: We couldn't connect to 'https://huggingface.co' - Host AI rule: 来源显示可能已有修复、规避或版本变化，说明书中必须标注适用版本。 - Why it matters: 可能增加新用户试用和生产接入成本。 - Evidence: community_evidence:github | https://github.com/postgresml/postgresml/issues/1682 | 来源讨论提到 docker 相关条件，需在安装/试用前复核。 - Hard boundary: 不要把这个坑点包装成已解决、已验证或可忽略，除非后续验证证据明确证明它已经关闭。