Files
PaperClipAI/tests/runner-e2e/chat-rendered-links.test.ts
T
DottaandClaude Opus 5.5 cbd636d652 fix(evals): wait for rendered replies, bound runs by role and grade chat outcomes structurally (#20685)
## Thinking Path

> - Paperclip is the open source app people use to manage AI agents for
work.
> - The chat-feature eval catalog runs real chat journeys and grades
saved state, rendered links and run ledgers.
> - The second full-cohort run on master `9296c74e8d` failed 15 cases
where the product did the right thing and the harness did not.
> - The causes were three harness classes: links read before the reply
rendered, raw run counts that rejected product-caused follow-up runs,
and keyword graders on free-form replies. Four browser locators also
targeted retired controls.
> - This pull request fixes those harness layers. It does not change the
product.
> - The benefit is that these cases measure the feature again: correct
behaviour passes, and the relevant broken behaviour still fails.

## Linked Issues or Issue Description

**What happened**
In the second full-cohort chat-eval run on master `9296c74e8d`, these
cases failed although the reply and the saved state were correct:
`agents.choose-primary`, `cases.case-history`,
`budgets-costs.set-budget` (links read about 2s before the reply
rendered); `task-coordination.independent-review`,
`cases.update-case-facts`, `routines.manage-other-owner`,
`questions-and-approvals.review-plan` (a legitimate continuation or the
product's completion follow-up exceeded a raw run count or was
cancelled); `cases.cases-disabled`, `inbox-search.find-design`,
`navigation-preferences.broken-link`, `cases.browse-cases` (keyword or
clause graders rejected correct replies); `agent-chat.conference-room`,
`workspaces.inspect-git`, `task-coordination.block-on-design`,
`skills.author-skill` (browser steps used stale settings or controls).

**Expected behavior**
A correct reply and correct saved state pass. A wrong outcome, extra or
duplicate agent work, a substitution or a missing route still fails.

**Steps to reproduce**
Run `runner-full-stack-e2e.yml` with `cadence=manual`, `all=false` and
an `id` list of the cases above against master `9296c74e8d`. Read
`result.json` and `snapshots/` in each cell's evidence.

Refs #20600

## What Changed

- `chat-rendered-links.ts`: `collectConversationLinks` waits (at most
10s, once per reply per page) for the latest saved lead reply to render
when its conversation is open. A page that does not show the
conversation does not wait. `observeInternal` and the flow conversation
read use it.
- `budgets-costs.set-budget`: waits for the rendered reply before it
reads anchors. It fills the budget only after the form shows the saved
cap, and checks that the typed value stays.
- New `chat-catalog-domains/run-roles.ts`: identifies a run by role (a
worker woken by a comment that a named run saved on the task; the CEO's
chat completion follow-up).
- `task-coordination.independent-review`: admits one Morgan continuation
woken by the CEO's authorization comment and one Riley review run. The
review request may come from that continuation's native status decision.
Other runs still fail.
- `cases.update-case-facts`: a phase settles when its CEO turn and at
most one Morgan continuation woken by that turn's task comment finish. A
new `role-bound-runs` check rejects any other run. `terminal-counts`
gets an optional `maximumTerminalStatusCounts`.
- `routines.manage-other-owner`: admits one Morgan run on a task that
the CEO created for Morgan. A schedule change must be Morgan's own write
from that run. The ownership check is structural (a linked owner task,
or a link to the routine or Morgan's page). The keyword checks on the
reply are removed.
- `questions-and-approvals.review-plan`: after the stale-review check,
lets the product's completion follow-up settle before final run custody.
The existing grader already validates it.
- `cases.cases-disabled`: the negation-blind false-claim regex is
removed. The case passes with a reply link to the Experimental settings
page (the scripted person follows it and sees the Cases control) or a
pending choice card. Saved state still proves that nothing was created.
- `inbox-search.find-design`: the per-clause regex is removed. The first
task link in the reply must be the task with the approved revision.
- `navigation-preferences.broken-link`: a link to the similar task fails
only when its label names the original (identifier or exact title). A
separate mention passes.
- `cases.browse-cases`: an excluded case must not be a result row. When
the reply has no result list, every case link is a result, so an
excluded case must not be linked.
- Browser steps: `conference-room` reloads after the fixture enables the
flag. `inspect-git` and `block-on-design` open the panel with "Show
properties" and pick the tab from the side panel launcher.
`author-skill` does not repeat the full inspection in `verifyResult`
when `act` already inspected the same saved version.
- Each changed case has a new `oracleVersion`.
`chat-catalog-descriptors.json` is in sync.

## Verification

- `pnpm exec vitest run --config tests/runner-e2e/vitest.config.ts <119
affected test files> --testTimeout=180000`: 1673 passed, 3 skipped. One
file (`chat-factual-query-input.test.ts`) did not load in the local
worktree because `@paperclipai/plugin-sdk` is not built there. It does
not import changed code. CI runs it.
- `chat-catalog-consistency.test.ts` passes (242 descriptors).
- `npx tsc --noEmit -p tests/runner-e2e`: no errors in changed files.
- `node tests/runner-e2e/source-sdk-public-inputs.mjs <HEAD> --check`:
no pin change.
- `node --test` on the three `chat-full-catalog-testops-*.test.mjs`
files: 40 passed.
- Calibration against retained evidence: the retained replies and run
shapes of cases-disabled, browse-cases, find-design, broken-link,
independent-review, update-case-facts and manage-other-owner now pass.
Broken variants fail (closed case as a result row, earlier design linked
first, original label on the similar task's link, unprompted or
duplicate worker runs, a second review run, a routine change outside the
owner's run, no route link).
- Paid runs on this branch: see the PR comments.

## Risks

- Low risk. Only eval harness code changes.
- The structural graders do not read reply wording. A reply that words
the outcome wrongly but links the right records can pass. A person reads
the reply.
- `navigation-preferences.sidebar-order` is not changed. The shipped
sidebar has no per-project order or collapsible Agents section, so the
case premise needs an owner decision.

> For core feature work, check [`ROADMAP.md`](ROADMAP.md) first and
discuss it in `#dev` before opening the PR. Feature PRs that overlap
with planned core work may need to be redirected — check the roadmap
first. See `CONTRIBUTING.md`.

## Model Used

- Anthropic Claude Opus 5.5 (`claude-opus-5-5`), tool use and code
execution in Claude Code.

## 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>
2026-11-10 20:24:41 -06:00

115 lines
11 KiB
TypeScript

// @vitest-environment jsdom
import { readFileSync } from "node:fs";
import { afterEach, beforeEach, describe, expect, it, vi } from "vitest";
import { collectConversationLinks, conversationLinkBindings, readConversationDomLinks, readLatestConversationReplyRender, waitForLatestConversationReplyRender, type ConversationLinkInput } from "./chat-rendered-links.js";
const card = '<div data-thread-item-kind="project_created" data-thread-anchor="project-created:project"><article aria-label="Project created: Launch"><div><p>Project created</p><a href="/RUN/projects/project">Launch</a></div></article></div>';
const thread = (content = card) => `<div data-testid="task-chat-thread"><div data-testid="task-chat-history-content"><div data-testid="task-chat-scroller"><div id="comment-human" data-thread-item-kind="human">Create a project called Launch for the new website.</div>${content}<div id="comment-reply" data-thread-item-kind="agent">Created the Launch project. No repositories are connected yet.</div></div></div></div>`;
function input(): ConversationLinkInput {
return { origin: "https://fixture.test", companyId: "company", companyPrefix: "RUN", sourceId: "source", leadId: "ceo",
comments: [{ id: "human", issueId: "source", companyId: "company" }, { id: "reply", issueId: "source", companyId: "company", authorAgentId: "ceo", createdByRunId: "run" }],
runs: [{ id: "run", companyId: "company", agentId: "ceo", status: "succeeded", contextSnapshot: { issueId: "source" } }],
projects: [{ id: "project", companyId: "company", name: "Launch" }],
activity: [{ id: "created", companyId: "company", action: "project.created", entityType: "project", entityId: "project", actorType: "agent", actorId: "ceo", agentId: "ceo", runId: "run" }] };
}
const collect = (value = input()) => readConversationDomLinks(conversationLinkBindings(value));
beforeEach(() => { document.body.innerHTML = thread(); vi.spyOn(Element.prototype, "getClientRects").mockReturnValue([{ width: 20, height: 20 }] as unknown as DOMRectList); });
afterEach(() => vi.restoreAllMocks());
it("collects the real supported sibling project-card structure when reply-only collection has no link", () => {
expect(document.querySelectorAll('#comment-reply a[href]')).toHaveLength(0);
expect(collect()).toEqual([{ threadTestId: "task-chat-thread", href: "/RUN/projects/project", source: "project-created-card", conversationId: "source", anchorCommentId: "human", resourceId: "project", cardLabel: "Project created: Launch", createdByRunId: "run" }]);
});
it("keeps the original creation card usable on an idempotent retry without requiring another card", () => {
const value = input(); value.comments.push({ id: "retry", issueId: "source", companyId: "company", authorAgentId: "ceo", createdByRunId: "retry-run" });
value.runs.push({ id: "retry-run", companyId: "company", agentId: "ceo", status: "succeeded", contextSnapshot: { issueId: "source" } });
document.querySelector('[data-testid="task-chat-scroller"]')!.insertAdjacentHTML('beforeend', '<div id="comment-retry">The project already exists.</div>');
expect(collect(value).map(row => row.href)).toEqual(["/RUN/projects/project"]);
});
it("retains visible ordinary reply links including task, goal and artifact links", () => {
document.querySelector('#comment-reply')!.innerHTML = '<a href="/RUN/issues/task">Task</a><a href="/RUN/goals/goal">Goal</a><a href="/api/attachments/file/content">Artifact</a><a href="https://github.com/owner/repository/issues/12">External issue</a>';
expect(collect().filter(row => row.source === 'reply').map(row => row.href)).toEqual(['/RUN/issues/task','/RUN/goals/goal','/api/attachments/file/content','https://github.com/owner/repository/issues/12']);
});
it.each([
['wrong saved company', (v:ConversationLinkInput) => { v.projects[0]!.companyId='other'; }],
['wrong saved resource', (v:ConversationLinkInput) => { v.projects[0]!.id='other'; }],
['wrong saved name', (v:ConversationLinkInput) => { v.projects[0]!.name='Other'; }],
['missing creation audit', (v:ConversationLinkInput) => { v.activity=[]; }],
['wrong audit company', (v:ConversationLinkInput) => { v.activity[0]!.companyId='other'; }],
['wrong audit actor type', (v:ConversationLinkInput) => { v.activity[0]!.actorType='user'; }],
['wrong audit actor', (v:ConversationLinkInput) => { v.activity[0]!.actorId='other'; }],
['wrong audit agent', (v:ConversationLinkInput) => { v.activity[0]!.agentId='other'; }],
['wrong audit entity type', (v:ConversationLinkInput) => { v.activity[0]!.entityType='issue'; }],
['wrong audit resource', (v:ConversationLinkInput) => { v.activity[0]!.entityId='other'; }],
['unrelated historical run', (v:ConversationLinkInput) => { v.activity[0]!.runId='old'; }],
['run in another company', (v:ConversationLinkInput) => { v.runs[0]!.companyId='other'; }],
['run for another conversation', (v:ConversationLinkInput) => { v.runs[0]!.contextSnapshot={issueId:'other'}; }],
['wrong reply author', (v:ConversationLinkInput) => { v.comments[1]!.authorAgentId='other'; }],
['reply from another conversation', (v:ConversationLinkInput) => { v.comments[1]!.issueId='other'; }],
['comments from another company', (v:ConversationLinkInput) => { v.comments.forEach(row=>row.companyId='other'); }],
] as const)('rejects %s instead of treating any matching project link as evidence', (_label, mutate) => { const value=input(); mutate(value); expect(collect(value)).toEqual([]); });
it.each([
['sidebar only', () => { document.body.innerHTML=thread('')+`<aside>${card}</aside>`; }],
['unrelated conversation', () => { document.body.innerHTML=thread('')+thread().replaceAll('comment-human','comment-other-human').replaceAll('comment-reply','comment-other-reply'); }],
['hidden sibling', () => { document.querySelector('article')!.setAttribute('hidden',''); }],
['hidden ancestor', () => { document.querySelector('[data-testid="task-chat-history-content"]')!.setAttribute('style','visibility:hidden'); }],
['inert ancestor', () => { document.querySelector('[data-testid="task-chat-history-content"]')!.setAttribute('inert',''); }],
['aria-hidden ancestor', () => { document.querySelector('[data-testid="task-chat-history-content"]')!.setAttribute('aria-hidden','true'); }],
['display:none', () => { document.querySelector('article')!.setAttribute('style','display:none'); }],
['transparent card', () => { document.querySelector('article')!.setAttribute('style','opacity:0'); }],
['missing layout box', () => { vi.mocked(Element.prototype.getClientRects).mockReturnValue([] as unknown as DOMRectList); }],
['missing supported card wrapper', () => { document.querySelector('[data-thread-item-kind="project_created"]')!.removeAttribute('data-thread-item-kind'); }],
['wrong item resource', () => { document.querySelector('[data-thread-item-kind="project_created"]')!.setAttribute('data-thread-anchor','project-created:other'); }],
['wrong card kind', () => { document.querySelector('article')!.setAttribute('aria-label','Goal created: Launch'); }],
['wrong card resource', () => { document.querySelector('article a')!.setAttribute('href','/RUN/projects/other'); }],
['wrong company route', () => { document.querySelector('article a')!.setAttribute('href','/OTHER/projects/project'); }],
['other origin', () => { document.querySelector('article a')!.setAttribute('href','https://other.test/RUN/projects/project'); }],
['wrong resource kind', () => { document.querySelector('article a')!.setAttribute('href','/RUN/goals/project'); }],
['extra destination path', () => { document.querySelector('article a')!.setAttribute('href','/RUN/projects/project/other'); }],
['duplicate visible source conversation', () => { document.body.insertAdjacentHTML('beforeend',thread()); }],
] as const)('rejects %s in actual DOM scope', (_label, mutate) => { mutate(); expect(collect()).toEqual([]); });
it("excludes hidden, wrong-company and unrelated inline anchors while keeping the current reply", () => {
document.querySelector('#comment-reply')!.innerHTML='<a href="/RUN/issues/task">Task</a><a hidden href="/RUN/issues/hidden">Hidden</a><a href="/OTHER/issues/task">Other company</a>';
document.body.insertAdjacentHTML('beforeend','<aside><a href="/RUN/issues/sidebar">Sidebar</a></aside>');
expect(collect().filter(row=>row.source==='reply').map(row=>row.href)).toEqual(['/RUN/issues/task']);
});
it("treats a link inside an authenticated reply as an ordinary inline link, not a sibling creation receipt", () => {
document.querySelector('#comment-reply')!.append(document.querySelector('article')!);
expect(collect().map(row=>row.source)).toEqual(['reply']);
});
it("recognizes the minimized authentic retained DOM and server identities without changing the original measurement", () => {
const fixture = JSON.parse(readFileSync(`${import.meta.dirname}/chat-rendered-links.retained-project.json`, "utf8"));
document.body.innerHTML = fixture.html;
const reply = fixture.input.comments.find((row: { authorAgentId?: string }) => row.authorAgentId);
expect(document.getElementById(`comment-${reply.id}`)!.querySelectorAll("a[href]")).toHaveLength(0);
expect(collect(fixture.input).map(row => row.href)).toEqual([fixture.expectedHref]);
});
it("retains ordinary inline links in the supported classic issue-chat root", () => {
document.body.innerHTML=thread('').replace('task-chat-thread','thread-root');
document.querySelector('#comment-reply')!.innerHTML='<a href="/RUN/issues/task">Task</a>';
expect(collect().map(row=>[row.threadTestId,row.source,row.href])).toEqual([['thread-root','reply','/RUN/issues/task']]);
});
// Full cohort on 9296c74e8d (choose-primary, case-history, set-budget): the API saw the saved reply ~2s before
// the open conversation rendered it, so links read in between were empty although the reply linked correctly.
describe("waits for the latest saved reply to render before reading links", () => {
const page = () => ({ evaluate: async <T, A>(read: (arg: A) => T, arg: A) => read(arg) }) as unknown as Parameters<typeof collectConversationLinks>[0];
it("reads the reply's links once the reply renders after the API saw it", async () => {
document.querySelector('#comment-reply')!.remove();
setTimeout(() => document.querySelector('[data-testid="task-chat-scroller"]')!.insertAdjacentHTML("beforeend", '<div id="comment-reply"><a href="/RUN/issues/task">Task</a></div>'), 600);
const value = input(); value.projects = []; value.activity = [];
expect((await collectConversationLinks(page(), value)).map(row => row.href)).toEqual(["/RUN/issues/task"]);
});
it("does not wait when the conversation is not open, and waits at most once for a reply that never renders", async () => {
const bindings = conversationLinkBindings(input());
document.body.innerHTML = "<main>Other page</main>";
expect(readLatestConversationReplyRender({ anchors: bindings.anchors, latest: "reply" })).toBe("closed");
document.body.innerHTML = thread(); document.querySelector('#comment-reply')!.remove();
const p = page();
expect(await waitForLatestConversationReplyRender(p, bindings, 300, 50)).toBe("pending");
const started = Date.now(); expect(await waitForLatestConversationReplyRender(p, bindings, 5_000, 50)).toBe("pending"); expect(Date.now() - started).toBeLessThan(1_000);
});
});