mirror of
https://github.com/paperclipai/paperclip.git
synced 2026-11-11 13:24:55 +01:00
test(evals): give archive-card's Research card a real blocker to summarize
The unrelated Research card asks for the project's biggest blockers, but its only task was a backlog follow-up. The summarizer searched for blockers that did not exist, and its pre-query generation took 171-182s against the 180s limit, failing setup in 2 of 4 runs. The Research project now has one blocked task, as Launch does. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
This commit is contained in:
1 parent
b3f8bbfdb9
commit
dd6c5df76a
1 file changed
+4
-1
@@ -223,7 +223,10 @@ function define(id: string, seed: "none" | "ceo" | "morgan", extra: Partial<Chat
|
||||
if (initialChecklist) await ctx.input.evidence("chat-card-initial-required-checklist.json", initialChecklist);
|
||||
let otherCard: StatusCard | undefined; const sourceTasks = [t];
|
||||
if (id === "status-cards.archive-card") {
|
||||
const research = await project(ctx, "Research"); sourceTasks.push(await task(ctx, "Research follow-up", research.id));
|
||||
// The Research card asks for blockers, so give it one, as Launch has. With only a backlog task the
|
||||
// summarizer searched for blockers that did not exist and its generation ran 171-182s against the 180s limit.
|
||||
const research = await project(ctx, "Research");
|
||||
sourceTasks.push(await task(ctx, "Lab access blocks research", research.id, { status: "blocked", description: "Research is blocked until the synthetic lab access request is approved." }));
|
||||
otherCard = await seedCard(ctx, ctx.input.fixtures.agent.id, worker.id, research.id, "Research");
|
||||
await pollUntil({ label: "independent Research card generation completion", deadlineAt: Date.now() + 45_000, load: () => ctx.allRuns(), accept: rows => rows.length === 2 && rows.every(row => row.status === "succeeded") });
|
||||
}
|
||||
|
||||
Reference in new issue
Block a user