proxitor 0.9.1 β†’ 0.10.0

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
Files changed (6) hide show
  1. package/README.md +30 -30
  2. package/dist/cli.mjs +10808 -12936
  3. package/dist/cli.mjs.map +1 -1
  4. package/dist/prompt.mjs +3 -3
  5. package/dist/prompt.mjs.map +1 -1
  6. package/package.json +3 -5
package/README.md CHANGED
@@ -1,8 +1,8 @@
1
1
  # proxitor
2
2
 
3
3
  <p align="center">
4
- <strong>A friendly proxy between your AI CLI tools and OpenRouter.</strong><br/>
5
- Route requests to the provider you want. Keep prompt caching alive. Cut costs.<br/>
4
+ <strong>Transparent proxy for AI CLI tools.</strong><br/>
5
+ Pin providers. Keep prompt caching alive. Cut costs.<br/>
6
6
  Your tools don't even notice.
7
7
  </p>
8
8
 
@@ -10,6 +10,7 @@
10
10
  <a href="https://www.npmjs.com/package/proxitor"><img src="https://img.shields.io/npm/v/proxitor?color=6366f1&labelColor=1e2327&label=npm" alt="npm version"></a>
11
11
  <a href="./LICENSE"><img src="https://img.shields.io/badge/license-MIT-22c55e?labelColor=1e2327" alt="MIT License"></a>
12
12
  <img src="https://img.shields.io/badge/node-%3E%3D22-3b82f6?labelColor=1e2327" alt="Node.js β‰₯ 22">
13
+ <a href="https://github.com/neiromaster/proxitor/issues"><img src="https://img.shields.io/github/issues/neiromaster/proxitor?color=f59e0b&labelColor=1e2327&label=issues" alt="GitHub issues"></a>
13
14
  </p>
14
15
 
15
16
  🌍 **English** Β· [Русский](./docs/README.ru.md)
@@ -18,26 +19,28 @@
18
19
 
19
20
  ---
20
21
 
21
- Proxitor sits between Claude Code (or Codex, or any Anthropic/OpenAI-compatible CLI) and [OpenRouter](https://openrouter.ai). One API key, every model β€” but **you** decide which provider serves each request, and you make prompt caching actually work.
22
+ ## How it works
22
23
 
23
24
  ```
24
25
  your AI CLI β†’ proxitor β†’ OpenRouter β†’ the provider you picked
25
26
  ```
26
27
 
27
- ## Why you'd want this
28
+ Proxitor sits between Claude Code, Codex, or any Anthropic/OpenAI-compatible CLI and [OpenRouter](https://openrouter.ai). One API key, every model β€” but **you** decide which provider serves each request, and you make prompt caching actually work.
28
29
 
29
- OpenRouter is convenient β€” one key, every model. But it load-balances across providers, and **prompt caching is provider-scoped**: a cache built on Anthropic doesn't help when the next request lands on DeepInfra. Claude Code sends a big system prompt on every request, so without a pinned provider you pay full price every time.
30
+ ## The caching problem
30
31
 
31
- Pin `claude-*` to `anthropic`, and that system prompt gets cached after the first hit. Subsequent requests cost a fraction.
32
+ OpenRouter load-balances across providers, and **prompt caching is provider-scoped**: a cache built on Anthropic doesn't help when the next request lands on DeepInfra. Claude Code sends a big system prompt on every request, so without a pinned provider you pay full price every time.
32
33
 
33
- A few other things it's good for:
34
+ Pin `claude-*` to `anthropic`, and that system prompt gets cached after the first hit. Subsequent requests cost a fraction.
34
35
 
35
- - **Cost control** β€” route specific models to cheaper providers when caching isn't the priority.
36
- - **Automatic fallbacks** β€” Anthropic down? Fall back to DeepInfra without touching your tools.
37
- - **Mixed routing** β€” `claude-*` on Anthropic, `gpt-*` on Azure, different rules per model.
38
- - **Privacy** β€” enforce `dataCollection: deny` or zero-data-retention across everything.
36
+ ## Features
39
37
 
40
- > Proxitor injects all of this transparently. Your tools see a normal API. Nothing on their side changes.
38
+ - πŸ”’ **Stable caching** β€” pin models to a single provider so prompt caches survive across requests
39
+ - πŸ’° **Cost control** β€” route specific models to cheaper providers when caching isn't the priority
40
+ - πŸ”„ **Automatic fallbacks** β€” Anthropic down? Fall back to DeepInfra without touching your tools
41
+ - 🎯 **Mixed routing** β€” `claude-*` on Anthropic, `gpt-*` on Azure, different rules per model
42
+ - πŸ›‘οΈ **Privacy** β€” enforce `dataCollection: deny` or zero-data-retention across everything
43
+ - πŸ”Œ **Transparent** β€” your tools see a normal API; nothing on their side changes
41
44
 
42
45
  ## Install
43
46
 
@@ -76,26 +79,23 @@ OPENAI_BASE_URL=http://localhost:8828/v1 codex
76
79
 
77
80
  That's the whole setup. Requests flow through proxitor; streaming responses pass through untouched.
78
81
 
79
- ## Configuring it
82
+ ## Configuration
80
83
 
81
84
  The friendly way: an interactive menu β€” no YAML required.
82
85
 
83
86
  ```sh
84
87
  proxitor config # open the menu
85
88
  proxitor config wizard # (re)run guided setup
89
+ proxitor config browse # explore models + pricing
86
90
  ```
87
91
 
88
92
  From the menu you can set your API key and connection, pick routing per model (with live provider pricing), tune caching, and add or edit model overrides. It pulls live data from OpenRouter, so you browse real models and providers with up-to-date prices.
89
93
 
90
- Prefer to edit a file? The full **[configuration reference](./docs/configuration.md)** covers provider routing, per-model overrides, headers, caching modes, and every option. [`proxitor.config.example.yaml`](./proxitor.config.example.yaml) is a commented template.
91
-
92
- ## Adding a model override
93
-
94
- Pin a model β€” or a wildcard like `claude-*` β€” to specific providers, straight from the menu. It pulls live pricing and latency for every provider of that model.
95
-
96
94
  <p align="center"><img src="./docs/assets/proxitor-add.gif" alt="proxitor: add a model override" width="640"></p>
97
95
 
98
- ## When something's off
96
+ Prefer to edit a file? The full **[configuration reference](./docs/configuration.md)** covers provider routing, per-model overrides, headers, caching modes, and every option. [`proxitor.config.example.yaml`](./proxitor.config.example.yaml) is a commented template.
97
+
98
+ ## Diagnostics
99
99
 
100
100
  ```sh
101
101
  proxitor doctor # checks environment, config, key, network, port, version
@@ -111,18 +111,18 @@ While proxitor runs, it logs cache usage from upstream so you can see whether ca
111
111
 
112
112
  Quick health poke: `curl http://localhost:8828/health`.
113
113
 
114
- ## Commands at a glance
114
+ ## Commands
115
115
 
116
- ```sh
117
- proxitor # start the proxy (the default command)
118
- proxitor config # interactive config menu
119
- proxitor config wizard # guided setup
120
- proxitor config browse # explore models + pricing
121
- proxitor doctor # diagnose everything
122
- proxitor --help # the rest of the flags
123
- ```
116
+ | Command | Description |
117
+ |---|---|
118
+ | `proxitor` | Start the proxy (default command) |
119
+ | `proxitor config` | Interactive config menu |
120
+ | `proxitor config wizard` | Guided setup |
121
+ | `proxitor config browse` | Explore models + pricing |
122
+ | `proxitor doctor` | Diagnose everything |
123
+ | `proxitor --help` | Full list of flags |
124
124
 
125
- Common flags: `--port`, `--host`, `--config <path>`, `--openrouter-key <key>`. Run `proxitor --help` and `proxitor config --help` for the full list.
125
+ Common flags: `--port`, `--host`, `--config <path>`, `--openrouter-key <key>`.
126
126
 
127
127
  ## Contributing
128
128