ur-agent 1.27.4 → 1.27.6
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/CHANGELOG.md +19 -0
- package/README.md +19 -9
- package/dist/cli.js +1955 -1159
- package/docs/CONFIGURATION.md +21 -2
- package/docs/USAGE.md +16 -5
- package/docs/providers.md +55 -54
- package/package.json +1 -1
package/CHANGELOG.md
CHANGED
|
@@ -1,5 +1,24 @@
|
|
|
1
1
|
# Changelog
|
|
2
2
|
|
|
3
|
+
## 1.27.6
|
|
4
|
+
|
|
5
|
+
- Route runtime requests through the selected provider/model pair instead of
|
|
6
|
+
allowing stale Ollama/default-provider paths to handle non-Ollama requests.
|
|
7
|
+
- Add runtime dispatch validation, backend labels, and focused mocked dispatch
|
|
8
|
+
tests for subscription CLI, API, Ollama, and OpenAI-compatible providers.
|
|
9
|
+
- Fix malformed message rendering from crashing startup when an undefined entry
|
|
10
|
+
appears in the UI message list.
|
|
11
|
+
|
|
12
|
+
## 1.27.5
|
|
13
|
+
|
|
14
|
+
- Make `/model` provider-first and provider-scoped, with clear subscription,
|
|
15
|
+
API-key, local runtime, and OpenAI-compatible server labels.
|
|
16
|
+
- Keep model discovery, validation, fallback, and saved config scoped to the
|
|
17
|
+
selected provider so CLI, API, and local model lists do not leak into each
|
|
18
|
+
other.
|
|
19
|
+
- Update provider documentation and tests for the provider/model selection
|
|
20
|
+
flow.
|
|
21
|
+
|
|
3
22
|
## 1.25.3
|
|
4
23
|
|
|
5
24
|
- Add provider alias resolution so `ur config set provider claude`,
|
package/README.md
CHANGED
|
@@ -143,9 +143,11 @@ ur --ollama-host http://192.168.1.50:11434
|
|
|
143
143
|
ur --discover-ollama
|
|
144
144
|
```
|
|
145
145
|
|
|
146
|
-
|
|
147
|
-
router over the configured Ollama host.
|
|
148
|
-
|
|
146
|
+
For Ollama sessions, model selection precedence is `OLLAMA_MODEL`, then
|
|
147
|
+
`UR_MODEL`, then the model router over the configured Ollama host. Once a
|
|
148
|
+
provider/model pair is selected with `/model` or `ur config set`, runtime
|
|
149
|
+
requests use that provider backend; they do not fall back to Ollama unless
|
|
150
|
+
`ollama` is the selected provider.
|
|
149
151
|
|
|
150
152
|
### Legal Provider Auth
|
|
151
153
|
|
|
@@ -179,19 +181,27 @@ Use quotes for shell values with spaces.
|
|
|
179
181
|
|
|
180
182
|
| Provider | Access type | Legal path |
|
|
181
183
|
| --- | --- | --- |
|
|
182
|
-
|
|
|
184
|
+
| Codex CLI | subscription | official Codex CLI login |
|
|
183
185
|
| Claude Code | subscription | official Claude Code login |
|
|
184
186
|
| Gemini CLI | subscription | official Gemini Code Assist login |
|
|
185
187
|
| Antigravity | subscription | official Antigravity login, where supported |
|
|
186
|
-
| OpenAI | API | `OPENAI_API_KEY` |
|
|
187
|
-
|
|
|
188
|
-
| Gemini | API | `GEMINI_API_KEY` |
|
|
188
|
+
| OpenAI API | API key | `OPENAI_API_KEY` |
|
|
189
|
+
| Claude API | API key | `ANTHROPIC_API_KEY` |
|
|
190
|
+
| Gemini API | API key | `GEMINI_API_KEY` |
|
|
189
191
|
| OpenRouter | API/router | `OPENROUTER_API_KEY` |
|
|
190
192
|
| Ollama | local | localhost Ollama runtime |
|
|
191
|
-
| LM Studio | local | local OpenAI-compatible server |
|
|
192
|
-
| llama.cpp | local | local OpenAI-compatible server |
|
|
193
|
+
| LM Studio | local/server | local OpenAI-compatible server |
|
|
194
|
+
| llama.cpp | local/server | local OpenAI-compatible server |
|
|
193
195
|
| vLLM | local/server | OpenAI-compatible server |
|
|
194
196
|
|
|
197
|
+
In the interactive app, `/model` shows providers first and then shows only the
|
|
198
|
+
selected provider's models. OpenAI API and Codex CLI, Claude API and Claude
|
|
199
|
+
Code, and Gemini API and Gemini CLI are separate access paths; subscription
|
|
200
|
+
logins and API keys are not interchangeable. The final confirmation includes
|
|
201
|
+
the runtime backend that will receive the next agent request. Use
|
|
202
|
+
`ur provider status` to show the active provider, model, access type, and
|
|
203
|
+
runtime backend.
|
|
204
|
+
|
|
195
205
|
Security policy: UR-AGENT never scrapes browser sessions, extracts OAuth
|
|
196
206
|
tokens, bypasses subscription/quota/region/organization restrictions, proxies a
|
|
197
207
|
consumer web session as an API, or claims support for a provider unless the
|