npm package discovery and stats viewer.

Discover Tips

  • General search

    [free text search, go nuts!]

  • Package details

    pkg:[package-name]

  • User packages

    @[username]

Sponsor

Optimize Toolset

I’ve always been into building performant and accessible sites, but lately I’ve been taking it extremely seriously. So much so that I’ve been building a tool to help me optimize and monitor the sites that I build to make sure that I’m making an attempt to offer the best experience to those who visit them. If you’re into performant, accessible and SEO friendly sites, you might like it too! You can check it out at Optimize Toolset.

About

Hi, 👋, I’m Ryan Hefner  and I built this site for me, and you! The goal of this site was to provide an easy way for me to check the stats on my npm packages, both for prioritizing issues and updates, and to give me a little kick in the pants to keep up on stuff.

As I was building it, I realized that I was actually using the tool to build the tool, and figured I might as well put this out there and hopefully others will find it to be a fast and useful way to search and browse npm packages as I have.

If you’re interested in other things I’m working on, follow me on Twitter or check out the open source projects I’ve been publishing on GitHub.

I am also working on a Twitter bot for this site to tweet the most popular, newest, random packages from npm. Please follow that account now and it will start sending out packages soon–ish.

Open Software & Tools

This site wouldn’t be possible without the immense generosity and tireless efforts from the people who make contributions to the world and share their work via open source initiatives. Thank you 🙏

© 2026 – Pkg Stats / Ryan Hefner

stormind-mcp

v1.17.1

Published

Stormind OS - OS Brain + MCP production agent. generate_document 一键成稿(compose→render→artifact→quality)+ security boundary(S1-S4)+ Agent Experience Boundary(Phase H)。

Readme

Stormind MCP Server v1.18.0

Stormind OS 的 MCP Server — Agent Harness 执行框架

这不是一个聊天机器人,也不是工具集合,而是一个可承载不同领域智能体的运行框架: 任何落在 os_core 上的智能体任务(学习规划 / 文件生成 / 项目规划 / 报告生产 / …)都经 理解 → 规划 → 选工具 → 执行 → 资产 → 质量 → 反馈 八层流水线统一编排与验证。

Agent Harness(Phase I)

harness_run 是八层执行框架的单一入口:

L1 能力边界 → L2 理解(intent + 信息不足追问)→ L3 规划(Task Graph)
→ L4 工具选择(权限门控 + 理由)→ L5 执行(状态/超时/重试/失败不可隐藏)
→ L6 资产(file_id 生命周期 created/active/expired/archived)
→ L7 质量(语义四维 + Quality Confidence)→ L8 反馈(经验库 + 下一次优化,namespace 隔离)

| Tool | 功能 | 输入 | 输出 | |------|------|------|------| | chat | 智能交互 + 理解层追问 | message | 文本 | | health_check | Provider 状态 | — | 文本 | | create_task_plan | 结构化执行计划 | goal / timeframe / context | Markdown(目标/阶段/时间线/Checklist/风险)| | create_document | 文档生成(md/txt)| title / content_spec / format | 文件 URL | | create_presentation | PPT 内容结构 | title / purpose / slides / style | PPT 大纲 | | create_image | 图片生成(预留)| prompt / style / size | 待 Agnes 多模态接入 | | create_word | Word 文档 | title / content / template | .docx(降级 md)| | create_pdf | PDF 文档 | title / content | .pdf(降级 txt)| | create_presentation_bin | 真实 PPT | title / slides / style | .pptx(降级大纲)| | generate_document | 高级生产 | request / output_format / template / style | 产物 + artifact + quality | | get_artifact | 下载产物 | file_id | Base64 + filename + MIME | | harness_run | Agent Harness 八层全链路 | message / scope / user / required_topics / stage | 理解+规划+工具+执行+资产+质量+反馈 |

办公能力(方案 C 分层降级)

| 依赖 | 有依赖 | 无依赖降级 | |------|--------|-----------| | python-docx | 真实 .docx | .md | | reportlab | 真实 .pdf | .txt | | python-pptx | 真实 .pptx | 大纲 .md |

依赖缺失不会导致工具失败,自动降级保持可用。

统一返回格式:

{ "status": "success", "tool": "create_document",
  "result": { "type": "md", "content": "", "url": "file://..." } }

能力

  • Provider:zhipu(主)、deepseek(备用)、agnes
  • 回退:主 Provider 失败(HTTP/超时/额度)自动切备用,stderr 记录
  • 协议:MCP stdio(JSON-RPC over stdin/stdout),零第三方依赖
  • 文件输出目录STORMIND_OUTPUT_DIR(默认 /tmp/stormind-output

启动

python3 mcp_server.py

环境变量

| 变量 | 必填 | 说明 | |------|:---:|------| | PROVIDER | — | 主 Provider,默认 zhipu | | ZHIPU_API_KEY | ✅(用 zhipu 时)| 智谱 API Key | | ZHIPU_MODEL | — | 默认 glm-4-flash | | FALLBACK_PROVIDER | — | 备用 Provider,如 deepseek | | DEEPSEEK_API_KEY | ✅(用 deepseek 时)| DeepSeek API Key | | DEEPSEEK_MODEL | — | 默认 deepseek-chat |

Key 只从环境变量 / STORMIND_ENV_FILE 指向的文件读取,不写代码、不打印。

产出物交付(Manifest + Base64,默认关闭)

云托管场景下 file:// 路径指向容器本地,用户拿不到文件本体。可选开启交付增强:

| 变量 | 默认 | 说明 | |------|:---:|------| | STORMIND_MANIFEST_FORMAT | off | json / yaml:每次产出在输出目录写 manifest_*.json/yaml(路径/类型/大小/摘要/时间戳),用户存本地做索引 | | STORMIND_EMBED_CONTENT | 0 | 1:二进制产出(docx/pdf/pptx)在响应里附 content_b64 + filename,用户 Base64 解码即得文件(不依赖云存储) | | STORMIND_EMBED_MAX_BYTES | 10485760 | Base64 内嵌大小上限(超过不内嵌,防超大响应) |

启用后响应 result 增加:filename / content_b64 / manifest。默认全关,行为与 v1.13 完全一致。

云托管说明

  • 不依赖任何本地绝对路径(/home/ubuntu/... 不必要)
  • 无本地 .env 文件时,自动回退读取平台注入的进程环境变量
  • 适用于讯飞星辰 MCP 云托管(command: python3 + args: mcp_server.py,或 npm 发布)

MCP stdio 测试

printf '%s\n' \
'{"jsonrpc":"2.0","id":1,"method":"initialize","params":{"protocolVersion":"2024-11-05","capabilities":{},"clientInfo":{"name":"t","version":"1"}}}' \
'{"jsonrpc":"2.0","method":"notifications/initialized"}' \
'{"jsonrpc":"2.0","id":2,"method":"tools/list","params":{}}' \
| python3 mcp_server.py 2>/dev/null