@smthrs/evals 1.0.0-rc.12 → 1.0.0-rc.14
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 +1 -1
- package/docs/api.md +1 -1
- package/docs/installation.md +2 -2
- package/package.json +5 -5
package/README.md
CHANGED
|
@@ -23,7 +23,7 @@ runner. Runtime evaluation code imports scorers directly.
|
|
|
23
23
|
Prereleases are published to npm under the `next` tag:
|
|
24
24
|
|
|
25
25
|
```bash
|
|
26
|
-
pnpm add @smthrs/evals@next effect
|
|
26
|
+
pnpm add @smthrs/evals@next effect@4.0.5
|
|
27
27
|
```
|
|
28
28
|
|
|
29
29
|
Requirements and entry points are at
|
package/docs/api.md
CHANGED
|
@@ -616,7 +616,7 @@ const aggregate = (cases: Readonly<Record<string, ReadonlyArray<boolean>>>, k: n
|
|
|
616
616
|
|
|
617
617
|
## A worked suite
|
|
618
618
|
|
|
619
|
-
[`
|
|
619
|
+
[`packages/smithers/agent/regression`](https://github.com/smithersai/smithers/tree/main/packages/smithers/agent/regression)
|
|
620
620
|
is a suite built on these modules that evaluates the Smithers agent from
|
|
621
621
|
[@smthrs/agent](/api/agent). Each case is a whole agent run against a scripted
|
|
622
622
|
model with no network access, reduced to one observation, and two scorers grade
|
package/docs/installation.md
CHANGED
|
@@ -7,13 +7,13 @@ Prereleases are published to npm under the `next` tag. Install the package
|
|
|
7
7
|
with its `effect` peer dependency:
|
|
8
8
|
|
|
9
9
|
```bash
|
|
10
|
-
pnpm add @smthrs/evals@next effect
|
|
10
|
+
pnpm add @smthrs/evals@next effect@4.0.5
|
|
11
11
|
```
|
|
12
12
|
|
|
13
13
|
## Runtime requirements
|
|
14
14
|
|
|
15
15
|
- Node.js 26.4.0 or later, from the package's `engines` field.
|
|
16
|
-
- `effect`
|
|
16
|
+
- `effect` `4.0.5`. Suites, runs, baselines, and gates are all `Effect`
|
|
17
17
|
values, so every program composes with the `effect` library directly.
|
|
18
18
|
- `@smthrs/core` supplies `Flow` values. `@smthrs/scorers` supplies scorers,
|
|
19
19
|
bindings, and the pure `@smthrs/scorers/ScoreGate` grading contract, including
|
package/package.json
CHANGED
|
@@ -1,7 +1,7 @@
|
|
|
1
1
|
{
|
|
2
2
|
"name": "@smthrs/evals",
|
|
3
3
|
"type": "module",
|
|
4
|
-
"version": "1.0.0-rc.
|
|
4
|
+
"version": "1.0.0-rc.14",
|
|
5
5
|
"license": "MIT",
|
|
6
6
|
"smthrs": {
|
|
7
7
|
"group": "agent"
|
|
@@ -165,11 +165,11 @@
|
|
|
165
165
|
"tag": "next"
|
|
166
166
|
},
|
|
167
167
|
"dependencies": {
|
|
168
|
-
"@smthrs/core": "1.0.0-rc.
|
|
169
|
-
"@smthrs/scorers": "1.0.0-rc.
|
|
168
|
+
"@smthrs/core": "1.0.0-rc.14",
|
|
169
|
+
"@smthrs/scorers": "1.0.0-rc.14"
|
|
170
170
|
},
|
|
171
171
|
"peerDependencies": {
|
|
172
|
-
"effect": "
|
|
172
|
+
"effect": "4.0.5"
|
|
173
173
|
},
|
|
174
174
|
"devDependencies": {
|
|
175
175
|
"@effect/language-service": "0.87.2",
|
|
@@ -177,7 +177,7 @@
|
|
|
177
177
|
"@types/node": "26.4.1",
|
|
178
178
|
"@vitest/coverage-v8": "5.0.0",
|
|
179
179
|
"dprint": "0.57.1",
|
|
180
|
-
"effect": "4.0.
|
|
180
|
+
"effect": "4.0.5",
|
|
181
181
|
"esbuild": "0.28.2",
|
|
182
182
|
"eslint": "9.39.5",
|
|
183
183
|
"eslint-import-resolver-typescript": "4.4.5",
|