P
PhongGuy/apinata
Native, local-first API client in Rust for HTTP, REST, GraphQL and SSE, with a desktop app, CLI and MCP server
在 GitHub 查看 ↗链接已于 2026-11-12 核查 · HTTP 200 · 打开的是厂商自己的官网
- ★ 数取值
- 1 ★ · PhongGuy/apinata · 取自 GitHub 搜索 API · (2026-11-13 20:00:50 核对)
- 来源明细
- 从该仓库的 GitHub 页面直接读取 · 条目类型:开源项目
- 收录与活跃
- 收录于 2026-11-12 · 最近见到 2026-11-14 · 共见到 3 次 · 被 2 个来源列出
- 评分构成
- 综合分 203 · 1 ★ · 2 个来源 · repo
常见问题(答案都来自我们的记录)
它适合谁 / 用来做什么?
我们把它归在这些标签下: 开源项目。标签来自条目自身,不是我们强加的分类。
谁维护它、还在更新吗?
仓库
PhongGuy/apinata;我们收录于 2026-11-12;最近见到 2026-11-14。这里没有"官方认证"或"质量评级"——我们只记录来源与时间。# Apinata
以下摘录自该仓库的 README(片段) · 在 GitHub 读完整 README →
Apinata is a native API client for HTTP, REST and GraphQL, written in Rust. It
also streams server-sent events. It is local-first: your requests are plain
YAML files on disk that live in git next to your code, with no account and no
cloud sync. There is no Electron and no
webview, so it starts fast and stays light. The same core drives a desktop app,
a command-line runner and an MCP server for AI assistants. It is free and open
source under MIT or Apache-2.0.
## Features
Collections are plain files. One YAML file per request, one folder per
folder, environments beside them, all readable in a diff and meant for git.
Changed requests are marked in the sidebar.
HTTP, GraphQL and server-sent events. GraphQL requests fetch the schema
by introspection, check queries against it and run subscriptions over
WebSocket. Event-stream responses show each event as it arrives. WebSocket
and gRPC request files appear in the sidebar but cannot be opened or sent
yet.
Many kinds of auth. Basic, Bearer, API key, Digest, AWS SigV4, NTLM,
OAuth 1.0, OAuth 2.0 (client credentials, password, authorization code
with PKCE, and implicit), WSSE and Akamai EdgeGrid. A cookie jar keeps
cookies between requests.
AI through MCP. An AI assistant can read and edit your collections and
send requests through the built-in MCP server, apinata mcp. Edits are plain
git-tracked files, and sending to a non-local host needs your approval in the
app, with a trusted-host list per collection.
Environments and secrets. Switch environments in one click. Secret
values stay in the system keychain, never in the collection files.
Tests, scripts and a CLI runner. Assertions and JavaScript scripts run
in the app and from apinata run, with JSON, JUnit and HTML reports for CI.
Scripts run before and after each request and can hold tests, with chai,
crypto and JWT helpers. Data-driven runs repeat a collection once per row of
a CSV or JSON file.
Import and export. Bring in collections from Postman (v2.0 and v2.1),
OpenAPI 3 and Swagger 2.0, Insomnia (v4 and v5), Hoppscotch and HAR 1.2,
environments from Postman and Hoppscotch, and single requests from curl.
Export to Postman v2.1, OpenAPI 3.1 (JSON or YAML) or one HTML page, and
record a run as HAR. Details are in
Importing and exporting.
Mock server and code generation. apinata mock serves your saved
examples, and apinata gen prints a request as code: curl, HTTPie, Python
requests, JavaScript fetch, Node axios, Go, PowerShell, PHP, Rust reqwest,
Java or C#.
## Download and install
The links below always download the latest release. To build it yourself
instead, see build from source.
| Platform | Download |
| Linux x86_64 | AppImage, .deb |
| Linux arm64 | AppImage, .deb |
| macOS Apple silicon | .dmg |
| macOS Intel | .dmg |
| Windows x64 | installer (.exe) |
Each release on the releases page
also carries the command-line runner on its own for every platform, and a
SHA256SUMS file. The .deb installs both apinata and apinata-app. Every
file in a release is listed in Release files.
### Signing and checksums
The macOS app is ad-hoc signed and not notarised, the Windows files are not
code-signed, and Linux has no OS signing; how this works is
also streams server-sent events. It is local-first: your requests are plain
YAML files on disk that live in git next to your code, with no account and no
cloud sync. There is no Electron and no
webview, so it starts fast and stays light. The same core drives a desktop app,
a command-line runner and an MCP server for AI assistants. It is free and open
source under MIT or Apache-2.0.
## Features
Collections are plain files. One YAML file per request, one folder per
folder, environments beside them, all readable in a diff and meant for git.
Changed requests are marked in the sidebar.
HTTP, GraphQL and server-sent events. GraphQL requests fetch the schema
by introspection, check queries against it and run subscriptions over
WebSocket. Event-stream responses show each event as it arrives. WebSocket
and gRPC request files appear in the sidebar but cannot be opened or sent
yet.
Many kinds of auth. Basic, Bearer, API key, Digest, AWS SigV4, NTLM,
OAuth 1.0, OAuth 2.0 (client credentials, password, authorization code
with PKCE, and implicit), WSSE and Akamai EdgeGrid. A cookie jar keeps
cookies between requests.
AI through MCP. An AI assistant can read and edit your collections and
send requests through the built-in MCP server, apinata mcp. Edits are plain
git-tracked files, and sending to a non-local host needs your approval in the
app, with a trusted-host list per collection.
Environments and secrets. Switch environments in one click. Secret
values stay in the system keychain, never in the collection files.
Tests, scripts and a CLI runner. Assertions and JavaScript scripts run
in the app and from apinata run, with JSON, JUnit and HTML reports for CI.
Scripts run before and after each request and can hold tests, with chai,
crypto and JWT helpers. Data-driven runs repeat a collection once per row of
a CSV or JSON file.
Import and export. Bring in collections from Postman (v2.0 and v2.1),
OpenAPI 3 and Swagger 2.0, Insomnia (v4 and v5), Hoppscotch and HAR 1.2,
environments from Postman and Hoppscotch, and single requests from curl.
Export to Postman v2.1, OpenAPI 3.1 (JSON or YAML) or one HTML page, and
record a run as HAR. Details are in
Importing and exporting.
Mock server and code generation. apinata mock serves your saved
examples, and apinata gen prints a request as code: curl, HTTPie, Python
requests, JavaScript fetch, Node axios, Go, PowerShell, PHP, Rust reqwest,
Java or C#.
## Download and install
The links below always download the latest release. To build it yourself
instead, see build from source.
| Platform | Download |
| Linux x86_64 | AppImage, .deb |
| Linux arm64 | AppImage, .deb |
| macOS Apple silicon | .dmg |
| macOS Intel | .dmg |
| Windows x64 | installer (.exe) |
Each release on the releases page
also carries the command-line runner on its own for every platform, and a
SHA256SUMS file. The .deb installs both apinata and apinata-app. Every
file in a release is listed in Release files.
### Signing and checksums
The macOS app is ad-hoc signed and not notarised, the Windows files are not
code-signed, and Linux has no OS signing; how this works is
收录徽章(可选)
想把收录标记放进 README 就复制这行,可选,不影响收录。
[](https://anclum.com/zh/s/phongguy-apinata/)
- 来源
- github · https://github.com/PhongGuy/apinata
- 按领域浏览
- 开源项目
- Traction
- 1 ★
- 收录来源
- 2 个独立来源
- 收录于
- 今天
同类推荐
所属领域与标签
每个标签都能点进去看同类的全部条目。数字是当前库里的真实条数。