@taskless/cli 0.8.0 → 0.9.0
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 -0
- package/dist/index.js +5869 -4720
- package/package.json +5 -2
package/README.md
CHANGED
|
@@ -176,6 +176,30 @@ All commands output structured JSON to stdout by default. Parse with `JSON.parse
|
|
|
176
176
|
|
|
177
177
|
## Developing
|
|
178
178
|
|
|
179
|
+
### Testing
|
|
180
|
+
|
|
181
|
+
```bash
|
|
182
|
+
pnpm --filter @taskless/cli test # run the suite once
|
|
183
|
+
pnpm --filter @taskless/cli exec vitest # watch mode
|
|
184
|
+
```
|
|
185
|
+
|
|
186
|
+
The suite runs entirely locally under vitest — no network, no auth, no agent.
|
|
187
|
+
Integration tests that exercise the built binary (for example `detect`) run
|
|
188
|
+
against `dist/`, so run `pnpm --filter @taskless/cli build` first (or after any
|
|
189
|
+
source change) before invoking them directly.
|
|
190
|
+
|
|
191
|
+
**Two kinds of test, one of which is not fully automatable.** Most tests are
|
|
192
|
+
deterministic unit/integration checks. The route-honesty dataset
|
|
193
|
+
(`test/fixtures/route-eval.json`) is different: the actual routing decision is
|
|
194
|
+
made by an _agent_ following `help/route.txt`, so it cannot be asserted by a
|
|
195
|
+
code classifier. The automated test (`test/route-eval.test.ts`) therefore only
|
|
196
|
+
**guards the dataset** — that it stays structurally valid and balanced across
|
|
197
|
+
every route and both failure directions (over-claim / over-escalate). Running
|
|
198
|
+
the dataset _as an evaluation_ — feeding each case to an agent and scoring its
|
|
199
|
+
chosen destination — is a separate, manual calibration step with more setup; it
|
|
200
|
+
is not part of `pnpm test`. Keep the two distinct: the suite proves the fixtures
|
|
201
|
+
are well-formed; an agent run proves the recipe routes honestly.
|
|
202
|
+
|
|
179
203
|
### API base URL
|
|
180
204
|
|
|
181
205
|
The CLI resolves the API base URL in this order:
|