@pickled-dev/cli 0.16.2 → 0.17.0
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 +34 -0
- package/dist/index.js +114 -112
- package/package.json +1 -1
package/README.md
CHANGED
|
@@ -162,6 +162,40 @@ Codebase loader safety defaults: skips directories (`onlyFiles`), does not follo
|
|
|
162
162
|
|
|
163
163
|
URL sources are NOT scanned by the audit's trap cross-reference in v1; they are fetched only during `pickled check`.
|
|
164
164
|
|
|
165
|
+
## Toolsets
|
|
166
|
+
|
|
167
|
+
Matrix mode (`scenario.matrix.toolsets`) iterates each scenario across named toolset profiles. v0.16.x ships two:
|
|
168
|
+
|
|
169
|
+
- **`none`** (the deterministic baseline). Pickled injects the cell's active source content into the agent's prompt. Citation contract applies if `requiredSources` is declared. Same scoring shape as non-matrix scenarios.
|
|
170
|
+
- **`web`** on Claude Code only. Maps to `allowedTools: ["WebSearch", "WebFetch"]` on the cell's Claude Code target. Source is NOT injected; the cell's prompt is rewritten to name the active source as the discovery target ("the canonical source for this question is at ..."). Citation contract is skipped; the cell scores on traps + `expected.includes`/`excludes`.
|
|
171
|
+
|
|
172
|
+
Declare profiles at the top level of `pickled.yml`:
|
|
173
|
+
|
|
174
|
+
```yaml
|
|
175
|
+
toolsets:
|
|
176
|
+
none: {}
|
|
177
|
+
web:
|
|
178
|
+
webSearch: true
|
|
179
|
+
webFetch: true
|
|
180
|
+
```
|
|
181
|
+
|
|
182
|
+
Then reference them per scenario:
|
|
183
|
+
|
|
184
|
+
```yaml
|
|
185
|
+
scenarios:
|
|
186
|
+
- name: "Install"
|
|
187
|
+
matrix:
|
|
188
|
+
interfaces: [quick]
|
|
189
|
+
sources: [llms]
|
|
190
|
+
toolsets: [none, web]
|
|
191
|
+
expected:
|
|
192
|
+
includes: ["bunx pickled"]
|
|
193
|
+
```
|
|
194
|
+
|
|
195
|
+
That scenario produces 2 cells: `[quick · llms · none]` (injected) and `[quick · llms · web]` (discovered via tools).
|
|
196
|
+
|
|
197
|
+
Custom toolset names that have no recognized adapter throw a clear "not yet implemented" error per cell. Web toolset on a non-Claude-Code interface throws "implemented only on the claude-code interface" so the misconfiguration is obvious.
|
|
198
|
+
|
|
165
199
|
## Targets
|
|
166
200
|
|
|
167
201
|
Pickled ships three target shapes today. Each target is a distinct surface that exercises the agent differently; results are comparable but not identical.
|