squad-kit 0.9.1 → 0.11.0
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.
- package/CHANGELOG.md +39 -0
- package/dist/cli.js +2871 -1283
- package/dist/console-ui/assets/index-BBs6jnF_.css +1 -0
- package/dist/console-ui/assets/index-n5R1pqsY.js +442 -0
- package/dist/console-ui/index.html +2 -2
- package/package.json +5 -4
- package/dist/cli.js.map +0 -1
- package/dist/console-ui/assets/index-C5I4cnED.css +0 -1
- package/dist/console-ui/assets/index-Dnb4Fisq.js +0 -416
package/CHANGELOG.md
CHANGED
|
@@ -7,6 +7,45 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
|
|
|
7
7
|
|
|
8
8
|
## [Unreleased]
|
|
9
9
|
|
|
10
|
+
## [0.11.0] — 2026-05-04
|
|
11
|
+
|
|
12
|
+
### What's new
|
|
13
|
+
|
|
14
|
+
- API `new-plan` can append planner events to `.squad/runs/<id>.events.jsonl`
|
|
15
|
+
(thinking text redacted on disk; retention rotates old files).
|
|
16
|
+
- Agent SDK runs emit richer events (tool lifecycle, thinking deltas, caps,
|
|
17
|
+
runtime metadata) for the CLI and console.
|
|
18
|
+
- API-mode `new-plan` prints stage pipeline, budget meters, and thinking
|
|
19
|
+
activity; the plan summary cites the events file path.
|
|
20
|
+
- `squad console` HTTP API lists persisted runs and serves stored event streams.
|
|
21
|
+
- Console UI adds a runs index, run report page, and a reworked Generate view
|
|
22
|
+
with live panels and navigation.
|
|
23
|
+
|
|
24
|
+
### Bug fixes
|
|
25
|
+
|
|
26
|
+
- `squad doctor` skips live external probes when `CI` is set and surfaces
|
|
27
|
+
planner probe errors cleanly.
|
|
28
|
+
- Vitest timeouts increased for slow config loads and console API tests.
|
|
29
|
+
|
|
30
|
+
## [0.10.0] — 2026-05-04
|
|
31
|
+
|
|
32
|
+
### What's new
|
|
33
|
+
|
|
34
|
+
- Anthropic planning defaults to the Claude Agent SDK so Opus 4.7 can use
|
|
35
|
+
adaptive thinking and effort; OpenAI and Google stay on the Vercel AI SDK.
|
|
36
|
+
- Planner YAML and CLI cover runtime choice, thinking mode, effort, and
|
|
37
|
+
per-phase effort (`--anthropic-runtime`, `--effort`, `--scout-effort`,
|
|
38
|
+
`--no-thinking`).
|
|
39
|
+
- The planner summary shows which runtime ran; run JSON may record runtime and
|
|
40
|
+
a provider-options snapshot for dashboards and tooling.
|
|
41
|
+
- `squad doctor` checks runtime wiring, model vs runtime fit, and the Agent SDK
|
|
42
|
+
binary.
|
|
43
|
+
- Published CLI no longer bundles `cli.js.map`; set `SQUAD_TSUP_SOURCEMAP=1`
|
|
44
|
+
when running `tsup` locally if you need a source map.
|
|
45
|
+
- On the Agent SDK path, cache token counts are not surfaced in telemetry
|
|
46
|
+
(caching still applies); live usage stays aggregate-only until the run ends;
|
|
47
|
+
cancel may not stop the upstream HTTP request immediately.
|
|
48
|
+
|
|
10
49
|
## [0.9.1] — 2026-05-04
|
|
11
50
|
|
|
12
51
|
### Bug fixes
|