@renxqoo/renx-code 0.0.8 → 0.0.9
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.
- package/README.md +103 -43
- package/package.json +1 -1
package/README.md
CHANGED
|
@@ -1,6 +1,6 @@
|
|
|
1
1
|
# Renx
|
|
2
2
|
|
|
3
|
-
`renx` is a terminal AI coding assistant.
|
|
3
|
+
`renx` is a terminal AI coding assistant for reading code, editing files, fixing errors, and working directly inside your current project.
|
|
4
4
|
|
|
5
5
|
- npm package: `@renxqoo/renx-code`
|
|
6
6
|
- command: `renx`
|
|
@@ -23,24 +23,24 @@ Start:
|
|
|
23
23
|
renx
|
|
24
24
|
```
|
|
25
25
|
|
|
26
|
-
|
|
26
|
+
## Quick Start
|
|
27
27
|
|
|
28
|
-
|
|
29
|
-
|
|
30
|
-
Run `renx` inside any project directory:
|
|
28
|
+
Open a project directory and run:
|
|
31
29
|
|
|
32
30
|
```bash
|
|
31
|
+
cd your-project
|
|
33
32
|
renx
|
|
34
33
|
```
|
|
35
34
|
|
|
36
35
|
`renx` uses the current terminal directory as the workspace by default.
|
|
37
36
|
|
|
38
|
-
Common things you can do:
|
|
37
|
+
Common things you can ask it to do:
|
|
39
38
|
|
|
40
|
-
-
|
|
41
|
-
-
|
|
42
|
-
-
|
|
43
|
-
-
|
|
39
|
+
- explain a codebase or a single file
|
|
40
|
+
- fix build or runtime errors
|
|
41
|
+
- modify files directly
|
|
42
|
+
- help write features, tests, and scripts
|
|
43
|
+
- investigate logs and terminal output
|
|
44
44
|
|
|
45
45
|
Built-in commands:
|
|
46
46
|
|
|
@@ -71,37 +71,19 @@ Default user data locations:
|
|
|
71
71
|
- `RENX_HOME/task/`
|
|
72
72
|
- `RENX_HOME/data.db`
|
|
73
73
|
|
|
74
|
-
|
|
75
|
-
|
|
76
|
-
Runtime:
|
|
77
|
-
|
|
78
|
-
- `RENX_HOME`
|
|
79
|
-
- `AGENT_MODEL`
|
|
80
|
-
- `AGENT_MAX_STEPS`
|
|
81
|
-
- `AGENT_MAX_RETRY_COUNT`
|
|
82
|
-
- `AGENT_TOOL_CONFIRMATION_MODE`
|
|
83
|
-
- `AGENT_CONVERSATION_ID`
|
|
84
|
-
- `AGENT_SESSION_ID`
|
|
85
|
-
|
|
86
|
-
Logging:
|
|
87
|
-
|
|
88
|
-
- `AGENT_LOG_LEVEL`
|
|
89
|
-
- `AGENT_LOG_FORMAT`
|
|
90
|
-
- `AGENT_LOG_CONSOLE`
|
|
91
|
-
- `AGENT_LOG_FILE_ENABLED`
|
|
92
|
-
|
|
93
|
-
File history:
|
|
74
|
+
Project config example:
|
|
94
75
|
|
|
95
|
-
|
|
96
|
-
-
|
|
97
|
-
|
|
98
|
-
- `AGENT_FILE_HISTORY_MAX_TOTAL_MB`
|
|
76
|
+
```text
|
|
77
|
+
your-project/.renx/config.json
|
|
78
|
+
```
|
|
99
79
|
|
|
100
|
-
|
|
80
|
+
Global config example:
|
|
101
81
|
|
|
102
|
-
|
|
82
|
+
```text
|
|
83
|
+
~/.renx/config.json
|
|
84
|
+
```
|
|
103
85
|
|
|
104
|
-
## Config Example
|
|
86
|
+
## Basic Config Example
|
|
105
87
|
|
|
106
88
|
```json
|
|
107
89
|
{
|
|
@@ -127,14 +109,92 @@ Provider API keys are still passed through their own environment variables, for
|
|
|
127
109
|
}
|
|
128
110
|
```
|
|
129
111
|
|
|
130
|
-
|
|
112
|
+
## Custom Model Config
|
|
131
113
|
|
|
132
|
-
|
|
133
|
-
|
|
114
|
+
You can define your own model in global or project `config.json` and then point `agent.defaultModel` to it.
|
|
115
|
+
|
|
116
|
+
Example:
|
|
117
|
+
|
|
118
|
+
```json
|
|
119
|
+
{
|
|
120
|
+
"agent": {
|
|
121
|
+
"defaultModel": "my-openai-compatible"
|
|
122
|
+
},
|
|
123
|
+
"models": {
|
|
124
|
+
"my-openai-compatible": {
|
|
125
|
+
"provider": "openai",
|
|
126
|
+
"name": "My OpenAI Compatible",
|
|
127
|
+
"baseURL": "https://example.com/v1",
|
|
128
|
+
"endpointPath": "/chat/completions",
|
|
129
|
+
"envApiKey": "MY_API_KEY",
|
|
130
|
+
"envBaseURL": "MY_API_BASE",
|
|
131
|
+
"model": "my-model",
|
|
132
|
+
"max_tokens": 8000,
|
|
133
|
+
"LLMMAX_TOKENS": 128000,
|
|
134
|
+
"features": ["streaming", "function-calling"]
|
|
135
|
+
},
|
|
136
|
+
"gpt-5.4": {
|
|
137
|
+
"baseURL": "https://my-proxy.example.com/v1"
|
|
138
|
+
}
|
|
139
|
+
}
|
|
140
|
+
}
|
|
134
141
|
```
|
|
135
142
|
|
|
136
|
-
|
|
143
|
+
This supports both:
|
|
137
144
|
|
|
138
|
-
|
|
139
|
-
|
|
145
|
+
- adding a completely new model
|
|
146
|
+
- overriding part of a built-in model, such as `baseURL`
|
|
147
|
+
|
|
148
|
+
## Environment Variables
|
|
149
|
+
|
|
150
|
+
Runtime:
|
|
151
|
+
|
|
152
|
+
- `RENX_HOME`
|
|
153
|
+
- `AGENT_MODEL`
|
|
154
|
+
- `AGENT_MAX_STEPS`
|
|
155
|
+
- `AGENT_MAX_RETRY_COUNT`
|
|
156
|
+
- `AGENT_TOOL_CONFIRMATION_MODE`
|
|
157
|
+
- `AGENT_CONVERSATION_ID`
|
|
158
|
+
- `AGENT_SESSION_ID`
|
|
159
|
+
|
|
160
|
+
Logging:
|
|
161
|
+
|
|
162
|
+
- `AGENT_LOG_LEVEL`
|
|
163
|
+
- `AGENT_LOG_FORMAT`
|
|
164
|
+
- `AGENT_LOG_CONSOLE`
|
|
165
|
+
- `AGENT_LOG_FILE_ENABLED`
|
|
166
|
+
|
|
167
|
+
File history:
|
|
168
|
+
|
|
169
|
+
- `AGENT_FILE_HISTORY_ENABLED`
|
|
170
|
+
- `AGENT_FILE_HISTORY_MAX_PER_FILE`
|
|
171
|
+
- `AGENT_FILE_HISTORY_MAX_AGE_DAYS`
|
|
172
|
+
- `AGENT_FILE_HISTORY_MAX_TOTAL_MB`
|
|
173
|
+
|
|
174
|
+
Custom models:
|
|
175
|
+
|
|
176
|
+
- `RENX_CUSTOM_MODELS_JSON`
|
|
177
|
+
|
|
178
|
+
Provider API keys are passed through their own environment variables, for example:
|
|
179
|
+
|
|
180
|
+
- `OPENAI_API_KEY`
|
|
181
|
+
- `GLM_API_KEY`
|
|
182
|
+
- `QWEN_API_KEY`
|
|
183
|
+
|
|
184
|
+
## Publish
|
|
185
|
+
|
|
186
|
+
Useful local release commands:
|
|
187
|
+
|
|
188
|
+
```bash
|
|
189
|
+
npm run pack:dry
|
|
190
|
+
npm run pack:tgz
|
|
191
|
+
npm run publish:patch
|
|
192
|
+
npm run publish:minor
|
|
193
|
+
npm run publish:major
|
|
140
194
|
```
|
|
195
|
+
|
|
196
|
+
Version rules:
|
|
197
|
+
|
|
198
|
+
- `publish:patch`: small fix, `0.0.1`
|
|
199
|
+
- `publish:minor`: new feature, `0.1.0`
|
|
200
|
+
- `publish:major`: breaking change or refactor, `1.0.0`
|