Files
PaperClipAI/tests/runner-e2e/screenshot-privacy.test.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

85 lines
6.7 KiB
TypeScript

import { execFileSync } from "node:child_process";
import { cp, mkdir, mkdtemp, readFile, rm, writeFile } from "node:fs/promises";
import { createHash } from "node:crypto";
import os from "node:os";
import path from "node:path";
import { afterEach, expect, it } from "vitest";
import { abnormalFinalizationResult } from "./abnormal-finalization.js";
import { runnerMatrix } from "./catalog.js";
import { regenerateRunnerDashboard } from "./dashboard-regenerate.js";
import { packageEvidence } from "./evidence.js";
import { publicScreenshotPaths, prunePrivateHistoryEvidence } from "./history-publish.js";
import { buildRunnerCampaign } from "./history.js";
import { sanitizePrivateReporterArtifacts } from "./private-reporter-artifacts.js";
import { validateRetainedRunnerResult } from "./result-validation.js";
import { screenshotWithholdingNotice, withholdPrivateScreenshots } from "./screenshot-privacy.js";
import type { RunnerE2EResult } from "./types.js";
const roots: string[] = [];
afterEach(async () => { await Promise.all(roots.splice(0).map(root => rm(root, { recursive: true, force: true }))); });
const png = Buffer.from("iVBORw0KGgoAAAANSUhEUgAAAAEAAAABCAQAAAC1HAwCAAAAC0lEQVR42mNk+A8AAQUBAScY42YAAAAASUVORK5CYII=", "base64");
function result(): RunnerE2EResult {
const e = runnerMatrix.find(row => row.suite.id === "chat-features" && row.profile.id === "runner-opencode" && row.environment.id === "local")!;
return { schema: "paperclip.runner-e2e.result/v2", executionId: e.id, suiteId: e.suite.id, suiteDefinitionHash: e.suiteDefinitionHash,
attempt: 1, status: "failed", failureClass: "candidate_failure", profileId: e.profile.id, environmentId: e.environment.id, caseId: e.task.id,
provider: e.profile.provider, model: e.profile.model, runtimeMode: e.profile.expectedRuntimeMode, startedAt: "2026-10-28T00:00:00.000Z",
finishedAt: "2026-10-28T00:00:01.000Z", durationMs: 1000, cleanup: "passed",
screenshots: [{ id: "failure", label: "Synthetic image", file: "failure.png", publication: "public-runner-fixture", sha256: createHash("sha256").update(png).digest("hex") }] };
}
it.each(["inner", "persisted", "reporter", "packaged"] as const)("retains %s privacy custody through cleanup and repeated clean finalization", finding => {
const original = result();
const held = withholdPrivateScreenshots(original, { innerSecretFinding: finding === "inner", persistedStateSecretFinding: finding === "persisted",
reporterEvidenceWithheld: finding === "reporter", packagedEvidenceSecretFinding: finding === "packaged" });
const wrapped = { ...abnormalFinalizationResult(held, new Error("packaging failed")), failureClass: "cleanup_failure" as const, cleanup: "failed" as const };
const final = withholdPrivateScreenshots(wrapped);
expect(final.screenshotPrivacy?.reasons).toHaveLength(1);
expect(final.screenshotPrivacy).toEqual(held.screenshotPrivacy);
expect(final.screenshots?.[0]).toEqual({ id: "failure", label: "Synthetic image", file: "failure.png", sha256: original.screenshots![0]!.sha256 });
expect(original.screenshots![0]!.publication).toBe("public-runner-fixture");
expect(final.status).toBe("failed"); expect(final.cleanup).toBe("failed");
expect(screenshotWithholdingNotice(final)).toContain("original images remain private");
expect(() => validateRetainedRunnerResult(final)).not.toThrow();
});
it("keeps clean behavior-failure screenshots publishable and rejects forged withholding metadata", () => {
const original = result();
expect(withholdPrivateScreenshots(original)).toBe(original);
expect(screenshotWithholdingNotice(original)).toBeNull();
const campaign = buildRunnerCampaign({ campaignId: "clean", generatedAt: original.finishedAt, expected: [original.executionId], results: [original] });
expect([...publicScreenshotPaths(campaign)]).toHaveLength(1);
for (const screenshotPrivacy of [{ withheld: false, reasons: ["inner-secret-finding"] }, { withheld: true, reasons: [] }, { withheld: true, reasons: ["arbitrary private text"] }])
expect(() => validateRetainedRunnerResult({ ...original, screenshotPrivacy })).toThrow("screenshotPrivacy");
});
it.each(["redacted", "omitted"] as const)("withholds synthetic PNG after actual reporter %s even when packaged scans are empty", async disposition => {
const root = await mkdtemp(path.join(os.tmpdir(), "screenshot-privacy-")); roots.push(root);
const privateDir = path.join(root, "private"), packaged = path.join(root, "packaged");
await mkdir(path.join(privateDir, "playwright-output"), { recursive: true });
await writeFile(path.join(privateDir, "failure.png"), png);
const secret = "synthetic-reporter-canary-only-2345";
if (disposition === "redacted") await writeFile(path.join(privateDir, "playwright-output", "error-context.md"), `Synthetic field ${secret}`);
else {
await mkdir(path.join(privateDir, "blob-report"));
await writeFile(path.join(root, "synthetic-report.txt"), secret);
execFileSync("zip", ["-q", path.join(privateDir, "blob-report", "report.zip"), "synthetic-report.txt"], { cwd: root });
}
const issues = await sanitizePrivateReporterArtifacts(privateDir, [secret]);
expect(issues).toMatchObject([{ disposition, reason: "exact secret value" }]);
const original = result(), held = withholdPrivateScreenshots(original, { reporterEvidenceWithheld: issues.length > 0 });
const final = { ...held, failureClass: "cleanup_failure" as const, cleanup: "failed" as const };
await writeFile(path.join(privateDir, "result.json"), JSON.stringify(final));
expect(await packageEvidence({ privateDir, uploadDir: packaged, secrets: [secret], expectPassScreenshot: false })).toMatchObject({ leaks: [], missing: [] });
const stage = path.join(root, "public-stage"), base = `evidence/${final.executionId}/attempt-1`;
await mkdir(path.join(stage, "evidence"), { recursive: true });
await cp(packaged, path.join(stage, base), { recursive: true });
const campaign = buildRunnerCampaign({ campaignId: "privacy-synthetic", generatedAt: final.finishedAt, expected: [final.executionId], results: [final] });
await writeFile(path.join(stage, "normalized-results.json"), JSON.stringify(campaign));
const allowed = publicScreenshotPaths(campaign);
expect([...allowed]).toEqual([]);
await prunePrivateHistoryEvidence(stage, allowed);
await expect(readFile(path.join(stage, base, "failure.png"))).rejects.toThrow();
expect(await readFile(path.join(privateDir, "failure.png"))).toEqual(png);
expect(await readFile(path.join(packaged, "failure.png"))).toEqual(png);
await regenerateRunnerDashboard({ bundle: stage });
const html = await readFile(path.join(stage, "index.html"), "utf8");
expect(html).toContain("Screenshots withheld after a privacy finding");
expect(html).not.toContain(`data-gallery-href="${base}/failure.png"`);
});