ur-agent 1.45.2 → 1.45.4
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 +32 -0
- package/README.md +10 -7
- package/dist/cli.js +3486 -2122
- package/docs/AGENT_FEATURES.md +20 -2
- package/docs/AGENT_TRENDS.md +3 -3
- package/docs/CONFIGURATION.md +11 -2
- package/docs/DEVELOPMENT.md +11 -0
- package/docs/TROUBLESHOOTING.md +9 -1
- package/docs/USAGE.md +19 -5
- package/docs/VALIDATION.md +27 -6
- package/documentation/app.js +47 -19
- package/documentation/index.html +22 -11
- package/extensions/jetbrains-ur/build.gradle.kts +1 -1
- package/extensions/vscode-ur-inline-diffs/package.json +1 -1
- package/package.json +1 -1
package/CHANGELOG.md
CHANGED
|
@@ -1,5 +1,37 @@
|
|
|
1
1
|
# Changelog
|
|
2
2
|
|
|
3
|
+
## 1.45.4
|
|
4
|
+
|
|
5
|
+
- Added mandatory provider-first model selection for the first interactive run
|
|
6
|
+
in every workspace that has no project-local model. The validated provider
|
|
7
|
+
and model pair is saved to `.ur/settings.local.json` before the REPL starts.
|
|
8
|
+
- Prevented user-global and built-in defaults from silently choosing a model
|
|
9
|
+
for a fresh folder. Explicit CLI/environment, shared project, flag, managed,
|
|
10
|
+
agent, and restored-session model choices continue without interruption.
|
|
11
|
+
- Made fresh headless workspaces fail before model execution with actionable
|
|
12
|
+
guidance to run the picker or pass `--model <model>`; initialization-only and
|
|
13
|
+
resume flows remain non-blocking. AutoApprove behavior is unchanged.
|
|
14
|
+
|
|
15
|
+
## 1.45.3
|
|
16
|
+
|
|
17
|
+
- Made slash-command resolution deterministic across bundled skills, plugins,
|
|
18
|
+
project skills, workflows, and built-ins. Duplicate canonical tokens are
|
|
19
|
+
rejected by source priority and conflicting aliases are removed; registry
|
|
20
|
+
tests now verify every shipped command token, description, and lazy loader.
|
|
21
|
+
- Removed overlapping `/paper`, `/security`, `/audit`, `/skills`, and
|
|
22
|
+
`/sandbox` registrations. The single `/sandbox` command now provides both
|
|
23
|
+
interactive settings and `status`, `check`, `init`, `eval`, and `exclude`
|
|
24
|
+
subcommands.
|
|
25
|
+
- Added `/ci-loop --cwd <path>` and working-directory evidence. Test-runner
|
|
26
|
+
"No tests found" failures now stop after one attempt with actionable cwd
|
|
27
|
+
guidance instead of invoking a fix agent repeatedly.
|
|
28
|
+
- Serialized concurrent artifact-viewer startup so simultaneous callers share
|
|
29
|
+
one server and one reported URL.
|
|
30
|
+
- Synchronized the user guides, static documentation site, and technical
|
|
31
|
+
specifications with the final command counts, source-priority rules,
|
|
32
|
+
`/skill` versus `/skills`, merged `/sandbox`, CI cwd behavior, and explicit
|
|
33
|
+
worktree publishing contract.
|
|
34
|
+
|
|
3
35
|
## 1.45.2
|
|
4
36
|
|
|
5
37
|
Correctness and containment release completing the runtime audit.
|
package/README.md
CHANGED
|
@@ -214,16 +214,19 @@ Development build detected. To update, pull latest source or install from npm.
|
|
|
214
214
|
Choose a model or Ollama host:
|
|
215
215
|
|
|
216
216
|
```sh
|
|
217
|
+
ur # first run opens provider/model selection
|
|
217
218
|
ur --model qwen2.5-coder:7b
|
|
218
219
|
ur --ollama-host http://192.168.1.50:11434
|
|
219
220
|
ur --discover-ollama
|
|
220
221
|
```
|
|
221
222
|
|
|
222
|
-
|
|
223
|
-
`
|
|
224
|
-
|
|
225
|
-
|
|
226
|
-
|
|
223
|
+
A fresh workspace never silently inherits a user-global or built-in model.
|
|
224
|
+
Interactive `ur` requires a provider/model choice and stores it in the
|
|
225
|
+
gitignored `.ur/settings.local.json`. Fresh print-mode runs require an explicit
|
|
226
|
+
`--model`, `OLLAMA_MODEL`, `UR_MODEL`, workspace setting, or managed setting.
|
|
227
|
+
For Ollama sessions, explicit environment precedence is `OLLAMA_MODEL`, then
|
|
228
|
+
`UR_MODEL`. Once selected, runtime requests use that provider backend; they do
|
|
229
|
+
not fall back to Ollama unless `ollama` is the selected provider.
|
|
227
230
|
|
|
228
231
|
### Legal Provider Auth
|
|
229
232
|
|
|
@@ -362,7 +365,7 @@ as first-class subcommands in the shipped CLI.
|
|
|
362
365
|
| `ur spec` | Default spec-first workflow: create requirements, design, and task documents under `.ur/specs/`, run the task list, and verify with strict proof gates. |
|
|
363
366
|
| `ur escalate` | Plan or run work with fast and oracle model tiers selected from local model capabilities. |
|
|
364
367
|
| `ur arena` | Run multiple agents on the same task in isolated worktrees and surface the winning diff. |
|
|
365
|
-
| `ur ci-loop` | Run a build or test command, hand failures to a fix agent, and retry with a bounded budget. |
|
|
368
|
+
| `ur ci-loop` | Run a build or test command in an explicit working directory, hand failures to a fix agent, and retry with a bounded budget. |
|
|
366
369
|
| `ur test-first` | Detect the project stack, run compile/test/lint commands, store failure traces, and install edit-time verify gates. |
|
|
367
370
|
| `ur safety` | Inspect or initialize project shell safety policy and evaluate command risk before execution. |
|
|
368
371
|
| `ur sandbox` | Inspect and manage the sandbox/permission architecture: status, dependency check, policy init, and command approval levels. |
|
|
@@ -556,7 +559,7 @@ ur context-pack remember --accepted "Use p-map for bounded concurrency" --scope
|
|
|
556
559
|
ur context-pack remember --rejected "Switch to esbuild" --alternative-to "Keep bun bundle"
|
|
557
560
|
ur context-pack remember --attempt "Tried Deno runtime" --status superseded
|
|
558
561
|
ur context-pack compress
|
|
559
|
-
ur ci-loop --command "bun test" --max-attempts 3 --dry-run
|
|
562
|
+
ur ci-loop --command "bun test" --cwd . --max-attempts 3 --dry-run
|
|
560
563
|
ur bg run "fix the flaky parser test" --worktree --dry-run
|
|
561
564
|
ur automation create nightly --schedule "0 9 * * 1-5" --prompt "Review open tasks"
|
|
562
565
|
ur repo-edit preview rename oldName --to newName
|