@brainbase-labs/cli 0.42.0 → 0.44.0
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/README.md +24 -1
- package/dist/index.js +1920 -1106
- package/package.json +1 -1
package/README.md
CHANGED
|
@@ -95,6 +95,19 @@ Both commands support `--json`. `send` exits 1 for unconfirmed, cancelled or
|
|
|
95
95
|
failed inputs, and for request errors. Other acknowledged states exit 0;
|
|
96
96
|
this confirms admission or delivery, not task success.
|
|
97
97
|
|
|
98
|
+
## Renaming and labelling tasks
|
|
99
|
+
|
|
100
|
+
Change a task's title or its metadata labels, including while it runs:
|
|
101
|
+
|
|
102
|
+
```sh
|
|
103
|
+
brainbase task update <task-id> --title "Nightly smoke run" --metadata env=staging
|
|
104
|
+
brainbase task update <task-id> --unset-metadata env --json
|
|
105
|
+
```
|
|
106
|
+
|
|
107
|
+
`--metadata` and `--unset-metadata` keep every label you do not name.
|
|
108
|
+
`--clear-metadata` removes them all, and `--clear-title` removes the title. See
|
|
109
|
+
[the command reference](docs/reference/commands.mdx) for every flag.
|
|
110
|
+
|
|
98
111
|
## Benchmark management
|
|
99
112
|
|
|
100
113
|
`benchmark init`, `benchmark validate`, and command help are local and need no
|
|
@@ -112,8 +125,13 @@ brainbase benchmark create ./benchmarks/support-quality --json
|
|
|
112
125
|
# A directory containing cases/*/files/** must be imported as a bundle.
|
|
113
126
|
brainbase benchmark import bundle ./benchmarks/support-quality --json
|
|
114
127
|
|
|
115
|
-
# Publish and
|
|
128
|
+
# Publish, then plan, start, and watch the run in one step. The run uses the
|
|
129
|
+
# agent's default model; --variant default=<model> runs another one.
|
|
116
130
|
brainbase benchmark publish <benchmark-id> --expected-version <version> --json
|
|
131
|
+
brainbase benchmark run <benchmark-id> --yes --watch --jsonl
|
|
132
|
+
brainbase benchmark run <benchmark-id> --yes --variant default=<model> --json
|
|
133
|
+
|
|
134
|
+
# Or plan and start separately.
|
|
117
135
|
brainbase benchmark run plan <benchmark-id> \
|
|
118
136
|
--revision <revision-id> \
|
|
119
137
|
--variant default \
|
|
@@ -280,6 +298,11 @@ providers. `default_model` can be changed with `brainbase agent push`; set it
|
|
|
280
298
|
to `null` to clear the override. Omitting either field leaves it unmanaged, so
|
|
281
299
|
older manifests keep their existing behavior.
|
|
282
300
|
|
|
301
|
+
When `agent create` claims a manifest with no `default_model`, the control plane
|
|
302
|
+
gives the agent the platform default for its harness, the same model the web
|
|
303
|
+
console picks, and `agent create` shows it. Cursor, Factory, and Qoder agents
|
|
304
|
+
get none, because those harnesses always run their vendor's own model.
|
|
305
|
+
|
|
283
306
|
## Feature support by harness
|
|
284
307
|
|
|
285
308
|
Each component type is stored in the place that harness already reads from at runtime — the CLI doesn't impose its own layout. "Native" cells describe where the install lands; "—" means the harness does not support that component type.
|