@lythos/skill-arena 0.9.8 → 0.9.10
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/package.json +1 -1
- package/src/runner.ts +13 -3
package/package.json
CHANGED
package/src/runner.ts
CHANGED
|
@@ -119,15 +119,25 @@ export async function runArenaFromToml(opts: {
|
|
|
119
119
|
criteria: [],
|
|
120
120
|
}) as JudgeVerdict
|
|
121
121
|
|
|
122
|
+
// Persist per-cell verdict + agent output for auditability
|
|
123
|
+
writeFileSync(join(cellDir, 'judge-verdict.json'), JSON.stringify({
|
|
124
|
+
...v,
|
|
125
|
+
agent_stdout: result.agentResult.stdout.slice(0, 5000),
|
|
126
|
+
agent_stderr: result.agentResult.stderr.slice(0, 1000),
|
|
127
|
+
duration_ms: result.agentResult.durationMs,
|
|
128
|
+
}, null, 2) + '\n')
|
|
129
|
+
|
|
122
130
|
if (!verdictsBySide.has(cell.side)) verdictsBySide.set(cell.side, [])
|
|
123
131
|
verdictsBySide.get(cell.side)!.push(v)
|
|
124
132
|
} catch (e) {
|
|
125
|
-
|
|
126
|
-
verdictsBySide.get(cell.side)!.push({
|
|
133
|
+
const errVerdict: JudgeVerdict = {
|
|
127
134
|
verdict: 'ERROR' as const,
|
|
128
135
|
reason: `Runner exception: ${e instanceof Error ? e.message : String(e)}`,
|
|
129
136
|
criteria: [],
|
|
130
|
-
}
|
|
137
|
+
}
|
|
138
|
+
writeFileSync(join(cellDir, 'judge-verdict.json'), JSON.stringify(errVerdict, null, 2) + '\n')
|
|
139
|
+
if (!verdictsBySide.has(cell.side)) verdictsBySide.set(cell.side, [])
|
|
140
|
+
verdictsBySide.get(cell.side)!.push(errVerdict)
|
|
131
141
|
}
|
|
132
142
|
}
|
|
133
143
|
|