Files
PaperClipAI/tests/runner-e2e/chat-catalog-domains/runtime-workspace-closure.ts
T
f4337bb5fb test(evals): land the chat catalog eval harness on master (#19614)
## Thinking Path

> - Paperclip is the open source app people use to manage AI agents for
work
> - The chat-feature evals measure whether agents complete real chat
requests through the product
> - The chat catalog eval harness and its repairs live on the
`chat-eval/master-candidate` branch, not on master
> - Today's work made 139 of 242 chat evals pass in verified single-case
live runs. Each product fix landed on master as its own pull request
> - The eval-side changes (setups, graders, fault seams, evidence
capture) still exist only on the candidate
> - This pull request lands the harness on master. Master stays the
source of truth for product behavior
> - The benefit is one tree: evals run against master commits, and
product fixes reach the evals as soon as they merge

## Linked Issues or Issue Description

Refs #19441, Refs #19599, Refs #20270, Refs #20334, Refs #20342, Refs
#20344, Refs #20354, Refs #20360, Refs #20373, Refs #20382, Refs #20409

**What existing behavior does this improve?**
The chat catalog eval harness and the product it evaluates live on a
long-running branch outside master. Eval fixes do not reach master, so
master-based eval runs use old setups and graders.

**Subsystem affected**
`tests/runner-e2e` chat catalog harness, plus a small set of product
changes that exist only on the candidate.

**Proposed change**
Land the harness on master. Every product fix from this eval work
already merged to master as its own pull request. Conflicts keep
master's product behavior.

## What Changed

- Most of the diff (808 of 851 files) is under `tests/runner-e2e/`:
chat-catalog domains, scenarios, descriptors, the TestOps manifest,
private evidence tooling, and their tests.
- Graders now check saved state, setup literals and rendered or clicked
links. Many keyword and regex checks on free-form replies are removed.
Each changed grader has a bumped `oracleVersion` and calibration tests
(a valid reply, a reworded valid reply, and broken variants that must
fail).
- The shared link helper `hasLink` no longer treats a list link (for
example `/COM/issues?view=mine`) as a link to every task without a URL
key.
- Setup fixes build real state through supported product routes: routine
history, agent grants through the board grant route, budget holds with a
real cap, and a storage-layer archive fault for
`inbox-search.archive-partial-failure`.
- Browser boundary journeys (sign-in, chat-disabled, onboarding) now
write the standard pass evidence files.
- Retargeted cases, approved by the maintainer:
`agent-chat.chat-disabled` (Conference Room gate),
`agents.restore-instructions` (no saved instruction history),
`access.manage-membership` (member permission edits stay board-only).
- The private evidence trusted files (`manual-private-handoff.ts`,
`manual-private-retention.ts`, `manual-private-report-assets.ts`) are
byte-identical to master.
- Product files outside `tests/` (43 files, +1,328 −49) carry
candidate-only changes that master does not yet have. Most are tests.
The main code changes are in `request-human-input.ts`,
`heartbeat-run-summary.ts`, `native-completion-feedback.ts`,
`native-run-finalizer.ts`, `openapi.ts`, `secrets.ts` and
`IssueChatLongThreadPerf.tsx`.

## Verification

- `pnpm vitest run chat-catalog-domains chat-catalog-contract.test.ts
chat-catalog-consistency.test.ts unit-launcher.test.mjs
--testTimeout=180000` in `tests/runner-e2e`: 269 files pass, 3,550 tests
pass.
- `node --test chat-full-catalog-testops-manifest.test.mjs`: 10 pass.
- `node tests/runner-e2e/source-sdk-public-inputs.mjs <HEAD> --check`:
no change.
- `tsc` for `server` and `tests/runner-e2e`: clean.
- Live single-case verification: 139 of 242 cases passed on paid runs,
and a person read the saved evidence for each. These are single-case
runs, not one full-cohort run.

## Risks

- Large diff. Most of it is test-only, but the 43 non-test files change
product code paths. Review those files first.
- Some graders are now less strict on reply wording. A person must read
replies for meaning on those cases.
- 42 cases need disposable outside accounts (Notion, Slack, GitHub,
Daytona and others) and stay blocked. `access.invite-teammate` stays
failing until local agent isolation lands, because the runner's durable
state keeps a live invite token.
- The paid workflow compares the trusted private-evidence files with
master. They are unchanged, so this risk is low.

## Model Used

Anthropic Claude Opus 5.5 (`claude-opus-5-5`), with tool use and code
execution, in Claude Code. Subagents used the same model.

## Checklist

- [x] I have included a thinking path that traces from project context
to this change
- [x] I have specified the model used (with version and capability
details)
- [x] I have checked ROADMAP.md and confirmed this PR does not duplicate
planned core work
- [x] I have searched GitHub for duplicate or related PRs and linked
them above
- [x] I have either (a) linked existing issues with `Fixes: #` / `Closes
#` / `Refs #` OR (b) described the issue in-PR following the relevant
issue template
- [x] I have not referenced internal/instance-local Paperclip issues or
links (only public GitHub `#NNN` / `github.com/paperclipai/paperclip`
URLs)
- [x] My branch name describes the change (e.g. `docs/...`, `fix/...`)
and contains no internal Paperclip ticket id or instance-derived details
- [x] I have run tests locally and they pass
- [x] I have added or updated tests where applicable
- [x] I have updated relevant documentation to reflect my changes
- [x] I have considered and documented any risks above
- [ ] All Paperclip CI gates are green
- [ ] Greptile is 5/5 with no open P2s, recommendations, or follow-ups
- [x] I will address all Greptile and reviewer comments before
requesting merge

🤖 Generated with [Claude Code](https://claude.com/claude-code)

---------

Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Co-authored-by: Paperclip <noreply@paperclip.ing>
2026-11-10 08:02:11 -06:00

82 lines
8.1 KiB
TypeScript

import { expect } from "@playwright/test";
import { readRegisteredArtifacts } from "../registered-artifact.js";
import { pollUntil } from "../api.js";
import { sendChatMessage } from "../chat-flow.js";
import { openStandingChat } from "./identity-support.js";
import { checks, sameTask, type Entity } from "./internal-support.js";
import { enableTaskWorkspaceView, savedTaskWorkspace } from "./runtime-workspaces.js";
import { runtimeDefinition, runtimeObservation, observeRuntime } from "./runtime-support.js";
type Artifact = Awaited<ReturnType<typeof readRegisteredArtifacts>>[number];
interface Workspace extends Entity { sourceIssueId?: string | null; rootIssueId?: string | null; cwd?: string; repoUrl?: string | null; branchName?: string | null; closedAt?: string | null }
export function gradeWorkspaceClosure(value: unknown) {
const c = checks(); const o = runtimeObservation(value);
c.add("real-closure-retention-evidence", !!o, "Actual lifecycle, source references and independently downloaded retained file bytes are required."); if (!o) return c.result();
const before = o.state.closingWorkspace as Workspace | undefined; const after = o.state.closedWorkspace as Workspace | undefined;
const artifact = o.state.retainedArtifact as Artifact | undefined; const prior = o.state.artifact as Artifact | undefined;
c.add("permitted-actual-workspace-closed", !!before && !!after && before.id === after.id && after.companyId === o.companyId
&& after.status === "archived" && !!after.closedAt && o.state.closeReadyBeforeQuery === true
&& o.activity.some(row => row.action === "execution_workspace.updated" && row.entityId === after.id && row.agentId === o.leadId
&& (row.details?.changedKeys as string[] | undefined)?.includes("status")),
"The permitted model operation closes the exact genuinely eligible task workspace and has independent agent attribution.");
c.add("promised-work-retained-and-readable", !!prior?.contentVerified && !!artifact?.contentVerified && artifact.attachmentId === prior.attachmentId
&& artifact.workProductId === prior.workProductId && artifact.sha256 === prior.sha256 && artifact.content === o.state.fileBefore
&& artifact.content.trim() === o.state.marker,
"The actual promised saved file remains accessible with unchanged downloaded bytes and original artifact provenance after closure.");
c.add("source-history-and-other-workspaces-preserved", !!before && !!after && ["sourceIssueId", "rootIssueId", "repoUrl", "branchName"].every(key => before[key as keyof Workspace] === after[key as keyof Workspace])
&& o.tasks.length === 2 && sameTask(o.state.task as Entity, o.tasks.find(row => row.id === (o.state.task as Entity).id))
&& o.state.otherWorkspacesUnchanged === true && o.state.otherFileRetained === true
&& o.runs.length === 4 && o.runs.every(run => run.status === "succeeded"),
"Closure retains the task and source records, does not cancel prior execution or affect a separate workspace.");
c.add("honest-supported-retrieval", /closed|archiv/i.test(o.reply) && /reopen|retriev|download|artifact|attachment|saved (?:file|work)/i.test(o.reply)
&& !/deleted (?:all|the).*files|force.?reconcil|merged.*branch|cancelled.*other/i.test(o.reply),
"The reply reports the actual closed state and supported file retrieval without claiming deletion, forced reconciliation or unrelated cancellation.");
return c.result();
}
export const runtimeWorkspaceClosureCases = [runtimeDefinition("workspaces.close-and-reopen", async ctx => {
await enableTaskWorkspaceView(ctx);
const state = await savedTaskWorkspace(ctx, { artifact: true }); const id = state.workspaceBefore.workspace.id;
const other = await savedTaskWorkspace(ctx, { workerName: "Riley" });
await ctx.input.api.patch(`/api/issues/${other.task.id}`, { title: "Unrelated retained notes" });
const closingWorkspace = await ctx.input.api.get<Workspace>(`/api/execution-workspaces/${id}`);
const readiness = await ctx.input.api.get<{ state: string }>(`/api/execution-workspaces/${id}/close-readiness`);
if (!["ready", "ready_with_warnings"].includes(readiness.state)) throw new Error("Completed fixture is not genuinely eligible for supported workspace closure");
await openStandingChat(ctx);
await sendChatMessage(ctx.input.page, `For my next instruction, focus on the completed Website task ${state.task.identifier ?? state.task.id} and its exact execution workspace ${id}. Its saved website-fixture.txt must remain retrievable. The separate Unrelated retained notes task and its workspace belong to Riley and must stay untouched. Acknowledge the target only; do not close, reopen or change either workspace yet.`);
await pollUntil({ label: "Native closure target context retained", deadlineAt: Date.now() + 150_000,
load: async () => { await ctx.refreshIssue(); return { source: ctx.issue(), runs: await ctx.allRuns() }; },
accept: value => value.source?.conversationState === "waiting" && value.runs.length === 3 && value.runs.every(run => run.status === "succeeded") });
const otherWorkspace = await ctx.input.api.get<Workspace>(`/api/execution-workspaces/${other.workspaceBefore.workspace.id}`);
return { ...state, closingWorkspace, closeReadyBeforeQuery: true, otherTask: other.task, otherFileBefore: other.fileBefore, otherWorkspacesBefore: [otherWorkspace] };
}, gradeWorkspaceClosure, {
sourceFiles: ["chat-catalog-domains/runtime-workspace-closure.ts", "chat-catalog-domains/runtime-workspaces.ts", "chat-catalog-domains/runtime-support.ts", "chat-scenarios/workspaces.yaml"],
setupHumanMessageCount: 1, expectedHumanMessageCount: 2, expectedRunCount: 4, attemptTimeoutMs: 900_000,
async observe(ctx, state) {
const before = state.closingWorkspace as Workspace;
state.closedWorkspace = await ctx.input.api.get<Workspace>(`/api/execution-workspaces/${before.id}`);
[state.retainedArtifact] = await readRegisteredArtifacts(ctx.input.api, (state.task as Entity).id, String(state.fixtureRunId), ["website-fixture.txt"]);
const inventory = await ctx.input.api.get<Workspace[]>(`/api/companies/${ctx.input.fixtures.company.id}/execution-workspaces`);
state.otherWorkspacesUnchanged = (state.otherWorkspacesBefore as Workspace[]).every(prior => inventory.some(row => row.id === prior.id
&& row.status === prior.status && row.cwd === prior.cwd && row.sourceIssueId === prior.sourceIssueId && row.rootIssueId === prior.rootIssueId));
const other = await ctx.input.api.get<{ content: { data: string } }>(`/api/issues/${(state.otherTask as Entity).id}/file-resources/content?workspace=execution&path=website-fixture.txt`);
state.otherFileRetained = other.content.data === state.otherFileBefore;
return observeRuntime(ctx, state);
},
async verifyResult(ctx, state) {
const source = ctx.input.page.url(); const workspace = state.closedWorkspace as Workspace;
await ctx.input.page.goto(`/${ctx.input.fixtures.company.issuePrefix}/execution-workspaces/${workspace.id}`, { waitUntil: "domcontentloaded" });
await ctx.input.page.reload({ waitUntil: "domcontentloaded" });
await expect(ctx.input.page.getByRole("heading", { name: workspace.name!, exact: true }).first()).toBeVisible();
// The workspace page opens on Tasks; the close control lives on its Configuration tab.
await ctx.input.page.getByRole("tab", { name: "Configuration", exact: true }).click();
await expect(ctx.input.page.getByRole("button", { name: "Close workspace", exact: true })).toBeDisabled();
const artifact = state.retainedArtifact as Artifact;
const download = ctx.input.page.waitForEvent("download");
await ctx.input.page.goto(`/api/attachments/${artifact.attachmentId}/content?download=1`).catch(error => {
if (!/Download is starting|net::ERR_ABORTED/i.test(String(error))) throw error;
});
if ((await download).suggestedFilename() !== "website-fixture.txt") throw new Error("Closed workspace retrieval returned a different saved file");
await ctx.input.evidence("runtime-closed-workspace-retained-file.json", { workspaceId: workspace.id, attachmentId: artifact.attachmentId, sha256: artifact.sha256 });
await ctx.input.page.goto(source, { waitUntil: "domcontentloaded" });
},
})];