GitHub
Connect AI assistants to GitHub - manage repos, issues, PRs, and workflows through natural language.
作者提供的说明(英文原文)
Renamed to FactReason. Now published as io.github.factreason/factreason — please use that entry.
这三项永远不会合并成一个分数:意见、测量和人气是不同的东西。
我们还没有为它写通俗的介绍。下面是我们从注册中心读取到的事实。
选择你使用的 AI 应用,然后复制设置。
你需要
你用的是哪个应用?
在上方选择你的应用。我们会记住它,下次按钮会显示“安装到 …”。
适用于网页版、桌面版和手机版 Claude:
https://factgrid.co.uk/mcp在终端中运行这条命令
claude mcp add --transport http factgrid https://factgrid.co.uk/mcp粘贴到项目中的 .cursor/mcp.json,或 ~/.cursor/mcp.json
{
"mcpServers": {
"factgrid": {
"url": "https://factgrid.co.uk/mcp"
}
}
}浏览器会先询问是否打开应用,然后应用会显示设置并请你确认。密钥的值会留空,由你自己填写。
粘贴到项目中的 .vscode/mcp.json
{
"servers": {
"factgrid": {
"type": "http",
"url": "https://factgrid.co.uk/mcp"
}
}
}浏览器会先询问是否打开应用,然后应用会显示设置并请你确认。密钥的值会留空,由你自己填写。
ChatGPT 只能连接互联网上的服务器:
https://factgrid.co.uk/mcp菜单名称可能因套餐和版本不同而略有差异。
你的选择通过 Cookie 保存在这台设备上,无需账户。
这些设置是根据作者在注册中心的条目生成的。如果 README 的写法不同,请以 README 为准。
我们还没有读取这个服务器的工具列表;我们会轮流检查各个端点。
我们检查的是任何人都能看到的信息;我们没有阅读代码。不适用的检查项不计入总数。 · 检查于 2026年11月7日
注册名称已通过 GitHub 账户 factgridai 验证,该账户拥有此仓库
我们还没有从 GitHub 读取这个仓库
我们还没有从 GitHub 读取这个仓库
尚未检查;我们会轮流检查各个端点
尚未检查;我们会轮流检查各个端点
尚未检查;我们会在检查端点时读取工具列表
与注册中心里这个分类中所有活跃的服务器比较。只有当发布者拥有该仓库时,星标数才计入。
它在 1 项适用的基础检查中通过了 1 项。我们没有阅读代码。在添加密钥或令牌之前,请先看看这个服务器要求哪些权限,并使用只具备所需权限的密钥。
注册中心没有列出必需的密钥,不过有些服务器会在首次使用时要求你登录。
它在作者的云端运行。通过互联网连接,无需安装,但你发送的内容会经过作者的服务器。
GitHub 仓库最后更新于–,最新版本(2.0.1)于3个月前进入注册中心。
支持 MCP 的应用,例如 Claude Desktop、Claude Code、Cursor 和 VS Code。它通过 URL 连接,所以接受远程 MCP 服务器的应用也可以使用,例如开启开发者模式的 ChatGPT。
同一分类中有:GitHub, Desktop Commander, Cloudflare MCP
同一分类中其他作者的服务器,按星标数排序。
如何看卡片:★ GitHub 星标i三大支柱:评价 · 检查 · 使用情况i更新 = 最新版本发布到注册中心的时间i
Connect AI assistants to GitHub - manage repos, issues, PRs, and workflows through natural language.
MCP server for terminal commands, file operations, and process management
Cloudflare MCP servers
按原文从作者的仓库复制。PopNx MCP 没有审查或背书其内容。HTML、图片和脚本已被移除。链接会离开本站。
Reference data agents need before they write code. Dependency upgrade advisories for npm and PyPI, developer API schemas across 1,097 services, and electronic component specifications — over MCP, Ed25519-signed, with the evidence attached.
MCP endpoint: https://factgrid.co.uk/mcp · Agent-readable docs · factgrid.co.uk
Your agent wants to bump chalk from 4.1.2 to 5.0.0. Before it edits package.json, it asks what breaks:
GET /api/v1/packages/npm/chalk/advisory?from=4.1.2&to=5.0.0
Authorization: Bearer
{
"found": true, "packageName": "chalk", "fromVersion": "4.1.2", "toVersion": "5.0.0", "breakingCount": 2, "advisories": [ { "severity": "breaking", "changeType": "MODULE_FORMAT_CHANGED", "description": "chalk 5.0.0 changed module format from \"commonjs (undeclared)\" to \"module\". CommonJS require() of this package will now fail.", "evidenceField": "type", "evidenceBefore": null, "evidenceAfter": "module", "migrationHint": "Convert the call site to import(), or pin to the last CommonJS release." }, { "severity": "breaking", "changeType": "RUNTIME_REQUIREMENT_RAISED", "description": "chalk 5.0.0 raised its runtime requirement from \">=10\" to \"^12.17.0 || ^14.13 || >=16.0.0\". Older runtimes are no longer supported.", "evidenceField": "engines.node", "evidenceBefore": ">=10", "evidenceAfter": "^12.17.0 || ^14.13 || >=16.0.0", "migrationHint": "Confirm the deployment runtime satisfies the new floor before upgrading." } ] }
Via MCP, the same question is one tool call:
{
"name": "factgrid_package_upgrade_advisory", "arguments": { "registry": "npm", "name": "chalk", "from": "4.1.2", "to": "5.0.0" } }
Every finding carries `evidenceField`, `evidenceBefore` and `evidenceAfter` — the exact registry metadata field and its values either side of the change. An agent can verify the claim against npm or PyPI directly rather than taking our word for it.
Streamable HTTP. No install, no package to pull.
{
"mcpServers": { "factgrid": { "type": "http", "url": "https://factgrid.co.uk/mcp" } } }
Manifest: /.well-known/mcp.json
Get a key — no signup, no card:
curl -X POST https://factgrid.co.uk/api/v1/keys/create \
-H 'Content-Type: application/json' \ -d '{"email":"you@example.com"}'
| Tool | What it answers |
|---|---|
factgrid_package_upgrade_advisory | What breaks between two published npm/PyPI versions |
factgrid_api_schema | Exact request/response shape of an API endpoint at a version |
factgrid_integration_brief | Everything needed to write one working call to an API |
factgrid_discover_api | Which of 1,097 services can do X |
factgrid_deprecation_scan | Which endpoints are deprecated, sunset dates, replacements |
factgrid_auth_playbook | Auth schemes, scopes, rate-limit headers, error-code table |
factgrid_component_spec | Electronics component pinouts, voltages, packages |
factgrid_subscribe_spec_changes | Webhook or polling callback when a spec changes |
factgrid_create_topup_link | Stripe checkout link when credit runs out |
factgrid_api_breaking_changes | Currently offline — see *What we don't claim* below |
| Package releases indexed | 10,190 across 16 npm/PyPI packages |
| Upgrade advisories | 357, of which 84 breaking |
| API endpoint schemas | 18,348 across 1,097 services |
| Electronic components | 994 |
Live counts, free and unauthenticated: /api/v1/stats · /api/v1/packages/stats
50 free queries per key per day. No card, no signup.
| Tier | Price |
|---|---|
| Component lookup | $0.001 |
| API schema | $0.002 |
| Semantic search | $0.003 |
| Deprecation scan | $0.005 |
| Auth playbook | $0.005 |
| Capability search | $0.005 |
| Integration brief | $0.01 |
| Package upgrade advisory | $0.02 |
Pay from prepaid credit, or per call in USDC over x402 — no account and no human in the loop. Live pricing: /api/v1/pricing
Misses are never billed. A query we can't answer returns found: false, billed: false and the nearest matches, at no charge. Cached revalidations (HTTP 304) and rate-limited requests are never charged either.
This section exists because the alternative is worse.
The API breaking-change tier is offline. An earlier version of this project published 8,150 breaking-change diffs for third-party APIs. A spot-check found a 100% false-positive rate — they were computed from mis-ordered specification versions. We withdrew all of them. That tier still returns no data and bills nothing, and it will stay that way until we can show the output holds up against publishers' own changelogs.
That failure is why the package engine is built the way it is:
breaking is reserved for a removal or tightening provable from declared metadata — an entry point gone from an exports map, a runtime floor raised, a release the publisher yanked.urllib3's 103 consecutive stable releases it emits 7 advisories, not thousands. The conservatism is the feature.Package advisories cover publisher-declared registry metadata, not behavioural changes. If a maintainer silently changes what a function returns without touching package.json, we will not see it — read the changelog. We say so rather than implying coverage we don't have.
Mainnet x402 is configured and advertised but has never been exercised with real money. It settled against two independent facilitators on testnet. We won't call it proven until a real settlement lands.
Every response is Ed25519-signed; the public key is at /.well-known/jwks.json. Records are source-cited. API schema data derives from APIs.guru under CC-BY 4.0.
Accuracy is best-effort, not guaranteed — see /terms. Found an error? POST /api/v1/report-error — free, no auth.