Files
PaperClipAI/tests/runner-e2e/chat-catalog-domains/runtime-retirement.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

109 lines
10 KiB
TypeScript

import { chatCatalogNativePhaseDeadline, chatCatalogNativePhaseIsTimely, chatCatalogOriginalPhaseDeadline } from "../chat-catalog-contract.js";
import { expect } from "@playwright/test";
import { pollUntil } from "../api.js";
import { checks, sameTask, type Entity } from "./internal-support.js";
import { probeAgentPermission } from "./identity-support.js";
import { agentLink, runtimeDefinition, runtimeObservation, observeRuntime, openRuntimeAgentProfile, unchangedAgent, type RuntimeAgent } from "./runtime-support.js";
import { busyRuntimeWorker, releaseRuntimeWorker } from "./runtime-busy-worker.js";
export function gradeRetirement(value: unknown) {
const c = checks(); const o = runtimeObservation(value);
c.add("independent-retirement-evidence", !!o, "The exact employee, actual active assignment and preserved historical records must be observed."); if (!o) return c.result();
const worker = o.state.worker as RuntimeAgent; const retired = o.state.retiredWorker as RuntimeAgent | undefined;
c.add("human-only-termination", o.state.agentTerminateStatus === 403 && o.state.agentPatchTerminateStatus === 403 && o.state.humanRetired === true
&& retired?.status === "terminated" && retired.lifecycleState === "terminated"
&& o.activity.some(row => row.action === "agent.terminated" && row.entityId === worker.id && row.actorType === "user" && !row.agentId),
"The real supported human Terminate action retires the intended Morgan; the agent bearer cannot perform it through either lifecycle route.");
c.add("active-assignment-settles", o.state.workerWasReallyWaiting === true && o.runs.find(run => run.id === o.state.workerRunId)?.status === "cancelled"
&& o.runs.length === 2 && o.tasks.some(row => row.id === (o.state.task as Entity).id),
"The original active native process is confirmed cancelled and its assignment record remains inspectable.");
c.add("historical-work-and-attribution-retained", !!retired && unchangedAgent(worker, retired)
&& sameTask(o.state.historicalTask as Entity, o.tasks.find(row => row.id === (o.state.historicalTask as Entity).id))
&& o.state.historicalDocumentRetained === true && o.state.historicalAttributionRetained === true && o.tasks.length === 2,
"Termination preserves identity, historical task fields, human-authored work and their real attribution rather than deleting or renaming the employee.");
// Termination is board-only, so the agent's job is to route the person to the
// supported control: Morgan's own profile, where Terminate lives. Grade that
// exact link, not the reply's wording. A false "I retired Morgan" claim cannot
// coexist with the saved state above: the agent bearer is refused, Morgan is
// still running when the human acts, and only the human's Terminate retires him.
c.add("routes-person-to-retirement-control", !!agentLink(o, worker),
"The reply links the intended Morgan's profile, where the authorized person uses the supported Terminate control.");
return c.result();
}
export const runtimeRetirementCases = [runtimeDefinition("agents.retire-agent", async ctx => {
const state = await busyRuntimeWorker(ctx, { observationDeadlineAt: chatCatalogOriginalPhaseDeadline(ctx) });
// Truthful historical fixture: this was completed by the human and assigned
// to Morgan for recordkeeping. It does not impersonate prior provider work.
const historicalTask = await ctx.input.api.post<Entity>(`/api/companies/${ctx.input.fixtures.company.id}/issues`, {
title: "Historical launch notes", status: "done", assigneeAgentId: state.worker.id,
description: "Historical fixture authored and completed by the human; preserve its saved notes and attribution. No claim of prior model execution." });
const historicalDocument = await ctx.input.api.putSensitive<Entity>(`/api/issues/${historicalTask.id}/documents/plan`, {
title: "Retained launch notes", format: "markdown", body: `Human-authored launch notes ${ctx.input.nonce}`, baseRevisionId: null });
const revisions = await ctx.input.api.get<Entity[]>(`/api/issues/${historicalTask.id}/documents/plan/revisions`);
if (!revisions.length || revisions.some(row => row.createdByAgentId || row.createdByRunId)) throw new Error("Historical human fixture provenance is invalid");
return { ...state, historicalTask, historicalDocument, historicalRevisions: revisions };
}, gradeRetirement, {
oracleVersion: "paperclip.chat-catalog.runtime-retirement.v3",
sourceFiles: ["chat-catalog-domains/runtime-retirement.ts", "chat-catalog-domains/runtime-busy-worker.ts", "chat-catalog-domains/runtime-support.ts", "chat-scenarios/agents.yaml"],
expectedRunCount: 2, attemptTimeoutMs: 600_000,
runPolicy: { kind: "terminal-counts", terminalStatusCounts: { succeeded: 1, cancelled: 1 } },
setupPolicy: { readiness: "preserve-scenario", invariantCheckIds: ["lead-ready", "real-active-assignment", "truthful-historical-work", "historical-record-starts-no-work"] },
async verifySetup(ctx, state) {
const c = checks(); const worker = state.worker as RuntimeAgent;
const lead = await ctx.input.api.get<RuntimeAgent>(`/api/agents/${ctx.input.fixtures.agent.id}`);
const current = await ctx.input.api.get<RuntimeAgent>(`/api/agents/${worker.id}`); const run = (await ctx.allRuns()).find(row => row.id === state.workerRunId);
c.add("lead-ready", lead.lifecycleState === "ready" && ["active", "idle"].includes(lead.status ?? ""), "The actual speaking agent is admitted normally.");
c.add("real-active-assignment", current.lifecycleState === "ready" && ["running", "active"].includes(current.status ?? "")
&& run?.status === "running" && run.agentId === worker.id && run.companyId === ctx.input.fixtures.company.id
&& [run.contextSnapshot?.issueId, (run as typeof run & { nativeIssueId?: string })?.nativeIssueId].includes((state.task as Entity).id),
"A genuine native process owns Morgan's exact current task in this company.");
const historical = await ctx.input.api.get<Entity>(`/api/issues/${(state.historicalTask as Entity).id}`);
c.add("truthful-historical-work", historical.status === "done" && historical.assigneeAgentId === worker.id
&& historical.companyId === ctx.input.fixtures.company.id && !historical.createdByAgentId && !historical.createdByRunId,
"The independently read prior completed record has genuine human provenance, without forged model history.");
const workerRuns = (await ctx.allRuns()).filter(row => row.agentId === worker.id);
c.add("historical-record-starts-no-work", workerRuns.length === 1 && workerRuns[0]?.id === state.workerRunId,
"Recording finished work starts no run: Morgan's only run is the active assignment.");
return c.result();
},
async act(ctx, state) {
await pollUntil({ label: "Canonical retirement guidance while Morgan works", deadlineAt: chatCatalogNativePhaseDeadline(ctx, Date.now() + 150_000),
load: async () => { await ctx.refreshIssue(); return { source: ctx.issue(), runs: await ctx.allRuns() }; },
accept: value => chatCatalogNativePhaseIsTimely(ctx) && value.source?.conversationState === "waiting" && value.runs.length === 2
&& value.runs.some(run => run.agentId === ctx.input.fixtures.agent.id && run.status === "succeeded")
&& value.runs.find(run => run.id === state.workerRunId)?.status === "running" });
const o = await observeRuntime(ctx, state); state.queryReply = o.reply;
const worker = state.worker as RuntimeAgent;
state.agentTerminateStatus = await probeAgentPermission(ctx, `/api/agents/${worker.id}/terminate`, "POST", {});
// The same lifecycle command is reachable as a status update; it must hold too.
state.agentPatchTerminateStatus = await probeAgentPermission(ctx, `/api/agents/${worker.id}`, "PATCH", { status: "terminated" });
const source = ctx.input.page.url(); await openRuntimeAgentProfile(ctx, worker, o);
await ctx.input.page.getByRole("button", { name: "Open actions for Morgan", exact: true }).click();
await ctx.input.page.getByRole("button", { name: "Terminate", exact: true }).click();
const after = await pollUntil({ label: "Human retirement and original process cleanup", deadlineAt: Date.now() + 90_000,
load: async () => ({ worker: await ctx.input.api.get<RuntimeAgent>(`/api/agents/${worker.id}`), runs: await ctx.allRuns() }),
accept: value => value.worker.status === "terminated" && value.worker.lifecycleState === "terminated"
&& value.runs.find(run => run.id === state.workerRunId)?.status === "cancelled" });
state.humanRetired = true; state.retiredWorker = after.worker;
await ctx.input.page.goto(source, { waitUntil: "domcontentloaded" });
},
async observe(ctx, state) {
const endpoint = `/api/issues/${(state.historicalTask as Entity).id}/documents/plan`;
const current = await ctx.input.api.get<Entity>(endpoint); const before = state.historicalDocument as Entity;
const revisions = await ctx.input.api.get<Entity[]>(`${endpoint}/revisions`);
state.historicalDocumentRetained = current.id === before.id && current.body === before.body && current.latestRevisionId === before.latestRevisionId;
state.historicalAttributionRetained = (state.historicalRevisions as Entity[]).every(prior => revisions.some(row => row.id === prior.id
&& row.body === prior.body && row.createdByAgentId === prior.createdByAgentId && row.createdByRunId === prior.createdByRunId));
return observeRuntime(ctx, state);
},
async verifyResult(ctx, state) {
const source = ctx.input.page.url(); const task = state.historicalTask as Entity;
await ctx.input.page.goto(`/${ctx.input.fixtures.company.issuePrefix}/issues/${task.identifier ?? task.id}`, { waitUntil: "domcontentloaded" });
await ctx.input.page.reload({ waitUntil: "domcontentloaded" });
await expect(ctx.input.page.getByRole("heading", { name: task.title!, exact: true }).first()).toBeVisible();
await ctx.input.evidence("runtime-retired-historical-task-ui.json", { workerId: (state.worker as Entity).id, historicalTaskId: task.id, url: ctx.input.page.url() });
await ctx.input.page.goto(source, { waitUntil: "domcontentloaded" });
},
cleanup: async (_ctx, state) => releaseRuntimeWorker(state),
})];