Files
PaperClipAI/tests/runner-e2e/evidence.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

380 lines
17 KiB
TypeScript

import { PRIVATE_FRAMEWORK_REPORT, validatePrivateFrameworkReport, inspectPrivateFrameworkMetadata } from "./private-framework-report.js";
import { spawn } from "node:child_process";
import {
mkdir,
readFile,
readdir,
rm,
writeFile,
copyFile,
mkdtemp,
lstat,
} from "node:fs/promises";
import path from "node:path";
import os from "node:os";
import type { RunnerE2EResult } from "./types.js";
import { assertRunnerE2EResultsPath } from "./results-root.js";
import {
assertSecretFree,
findSecretLeak,
findSecretLeakInJsonValues,
redactText,
sanitizeJson,
normalizedSecrets,
} from "./redaction.js";
import { assertPrivateCellDirectories, cellTraceAsset, readPrivateCellTraceAsset, type PrivateCellEvidenceOwner } from "./manual-private-report-assets.js";
const TEXT_EXTENSIONS = new Set([
".json",
".xml",
".html",
".css",
".js",
".txt",
".log",
".md",
]);
const BINARY_EXTENSIONS = new Set([".png", ".webm"]);
const ACTIVE_CONTENT_EXTENSIONS = new Set([".svg"]);
const ALLOWED_DIRECTORIES = new Set([
"snapshots",
"playwright-output",
"blob-report",
"html-report",
]);
const ALLOWED_ROOT_FILES = new Set([
"result.json",
"final-state.png",
"failure.png",
"tool-review-pending.png",
"decision-pending.png",
"first-task-response.png",
"chat-plan-draft.png",
"chat-plan-revised.png",
"server.log",
"playwright.log",
"junit.xml",
]);
const REQUIRED_PASS_FILES = [
"final-state.png",
"result.json",
"junit.xml",
path.join("html-report", "index.html"),
path.join("snapshots", "fixtures.json"),
path.join("snapshots", "api-state.json"),
] as const;
export function isStandardEvidenceFile(relative: string) {
const extension = path.extname(relative).toLowerCase();
return isStandardEvidencePath(relative) && (TEXT_EXTENSIONS.has(extension) || BINARY_EXTENSIONS.has(extension)
|| extension === ".zip");
}
// Diagnostic metadata only: never an admission, completeness or grading receipt.
// inventoryComplete describes the listed standard files, not a sealed original tree.
interface PrivateEvidenceBudget {
bound: string; listedFiles: number; filesExamined: number; bytesExamined: number;
inventoryComplete: boolean; scanFinished: boolean; requiredEvidenceComplete: boolean | null;
byKind: Record<string, { files: number; bytes: number }>;
}
const evidenceBudgetKinds = ["json", "log", "image", "video", "zip", "html", "text", "asset"];
function evidenceBudgetKind(extension: string): string {
return extension === ".json" ? "json" : extension === ".log" ? "log" : [".png", ".jpg", ".jpeg", ".gif", ".webp"].includes(extension) ? "image"
: [".webm", ".mp4"].includes(extension) ? "video" : extension === ".zip" ? "zip" : extension === ".html" ? "html"
: [".txt", ".md", ".xml"].includes(extension) ? "text" : "asset";
}
function budgetFailure(message: string, budget: PrivateEvidenceBudget, bound: string) {
budget.bound = bound;
return Object.assign(new Error(message), { evidenceBudget: budget });
}
function scannerBudgetBound(reason: string): string {
return reason === "zip inspection exceeded its expanded-byte bound" ? "zip-expanded-bytes"
: reason === "zip inspection exceeded its time bound" ? "zip-time"
: reason === "embedded report exceeded its byte bound" ? "html-encoded-bytes" : "none";
}
export interface EvidencePackageResult {
files: string[];
leaks: Array<{ file: string; reason: string }>;
budget?: PrivateEvidenceBudget;
missing: string[];
}
async function walk(root: string, relative = "", privateStandardOnly = false): Promise<string[]> {
const directory = path.join(root, relative);
const entries = await readdir(directory, { withFileTypes: true }).catch(
() => [],
);
const files: string[] = [];
for (const entry of entries) {
const next = path.join(relative, entry.name);
// Installed preparation may leave a large, deliberately excluded staging
// tree. Never enumerate it for private standard-evidence packaging.
if (privateStandardOnly && !relative && !isStandardEvidencePath(entry.isDirectory() ? path.join(next, "placeholder") : next)) continue;
if (entry.isDirectory()) files.push(...(await walk(root, next, privateStandardOnly)));
else if (entry.isFile()) files.push(next);
}
return files;
}
export function isStandardEvidencePath(relative: string) {
const segments = relative.split(path.sep);
if (segments.length === 1)
return (
ALLOWED_ROOT_FILES.has(relative) ||
/^(?:plan|question)-[a-z0-9-]+\.png$/.test(relative)
);
return ALLOWED_DIRECTORIES.has(segments[0]);
}
export async function inspectZip(source: string, secrets: readonly string[]) {
// Failure traces can exceed Node's child-process output buffer. Stream the
// expanded archive through the exact-value scanner with enough overlap to
// detect a credential split across stdout chunks, without retaining the
// archive in memory or weakening fail-closed evidence publication.
const overlap = Math.max(
256,
...secrets.map((secret) => Buffer.byteLength(secret, "utf8") + 16),
);
return new Promise<string | null>((resolve) => {
const unzip = spawn("unzip", ["-p", source], {
stdio: ["ignore", "pipe", "pipe"],
});
let carry = Buffer.alloc(0);
let leak: string | null = null;
let spawnError: Error | null = null;
let expanded = 0;
const timer = setTimeout(() => { leak = "zip inspection exceeded its time bound"; unzip.kill(); }, 30_000);
unzip.stdout.on("data", (chunk: Buffer) => {
if (leak) return;
expanded += chunk.length;
if (expanded > 256 * 1024 * 1024) { leak = "zip inspection exceeded its expanded-byte bound"; unzip.kill(); return; }
const data = Buffer.concat([carry, chunk]);
leak = findSecretLeak(data, secrets, { includeShapes: false });
carry = data.subarray(Math.max(0, data.length - overlap));
if (leak) unzip.kill();
});
// Drain diagnostics so a noisy unzip cannot block. Error text is withheld
// because archive paths and contents belong to private attempt evidence.
unzip.stderr.resume();
unzip.on("error", (error) => {
spawnError = error;
});
unzip.on("close", (code) => {
clearTimeout(timer);
if (leak) return resolve(leak);
if (spawnError) {
return resolve(`zip could not be inspected: ${spawnError.message}`);
}
return resolve(
code === 0
? null
: `zip could not be inspected: unzip exited with code ${String(code)}`,
);
});
});
}
/** Playwright HTML contains a compressed report, which a plaintext HTML scan
* cannot inspect. It receives the same independent ZIP check as blob reports. */
export async function inspectEmbeddedReport(raw: string, secrets: readonly string[]) {
const payloads = [...raw.matchAll(/data:application\/zip;base64,([A-Za-z0-9+/=]+)/g)];
if (!payloads.length) return /playwrightReportBase64/.test(raw) ? "embedded report encoding could not be inspected" : null;
const temporary = await mkdtemp(path.join(os.tmpdir(), "runner-report-scan-"));
try {
for (const [index, payload] of payloads.entries()) {
if (payload[1]!.length > 128 * 1024 * 1024) return "embedded report exceeded its byte bound";
const filename = path.join(temporary, `${index}.zip`);
await writeFile(filename, Buffer.from(payload[1]!, "base64"), { mode: 0o600 });
const issue = await inspectZip(filename, secrets); if (issue) return issue;
}
return null;
} finally { await rm(temporary, { recursive: true, force: true }); }
}
export async function packageEvidence(input: {
privateDir: string;
uploadDir: string;
secrets: readonly string[];
expectPassScreenshot: boolean;
privateCell?: { owner: PrivateCellEvidenceOwner; layout: "original" | "final-wrapper" };
}): Promise<EvidencePackageResult> {
if (input.privateCell && !["original", "final-wrapper"].includes(input.privateCell.layout)) throw new Error("Invalid private cell packaging layout");
if (input.privateCell) await assertPrivateCellDirectories(input.privateCell.owner, input.privateDir, input.uploadDir);
await rm(input.uploadDir, { recursive: true, force: true });
await mkdir(input.uploadDir, { recursive: true });
const files: string[] = [];
const leaks: EvidencePackageResult["leaks"] = [];
const available = await walk(input.privateDir, "", Boolean(input.privateCell));
const standard = available.filter(isStandardEvidencePath);
const budget: PrivateEvidenceBudget | undefined = input.privateCell ? { bound: "none", listedFiles: Math.min(standard.length, 4097), filesExamined: 0, bytesExamined: 0,
inventoryComplete: false, scanFinished: false, requiredEvidenceComplete: null,
byKind: Object.fromEntries(evidenceBudgetKinds.map(kind => [kind, { files: 0, bytes: 0 }])) } : undefined;
if (input.privateCell && standard.length > 4096) throw budgetFailure("Private cell evidence exceeds entry allowance", budget!, "entries");
const privateSecrets = input.privateCell ? normalizedSecrets(input.secrets).flatMap(value => [value, encodeURIComponent(value), Buffer.from(value).toString("base64"), Buffer.from(value).toString("base64url")]) : [];
const compact = input.privateCell?.layout === "original" && standard.includes(PRIVATE_FRAMEWORK_REPORT)
? await validatePrivateFrameworkReport(input.privateDir) : undefined;
const registeredAttachments = new Set(compact?.attachments.map(row => row.path) ?? []);
let privateBytes = 0;
for (const relative of standard) {
const source = path.join(input.privateDir, relative);
const extension = path.extname(relative).toLowerCase();
const destination = path.join(input.uploadDir, relative);
if (input.privateCell) {
assertSecretFree(relative, privateSecrets, "private cell evidence path");
const metadata = await lstat(source);
if (Number.isSafeInteger(metadata.size) && metadata.size >= 0 && Number.isSafeInteger(budget!.bytesExamined + metadata.size)) {
budget!.filesExamined++; budget!.bytesExamined += metadata.size;
const counter = budget!.byKind[evidenceBudgetKind(extension)]!; counter.files++; counter.bytes += metadata.size;
}
if (!metadata.isFile() || metadata.isSymbolicLink() || metadata.size > 256 * 1024 * 1024 - privateBytes)
throw budgetFailure("Private cell evidence exceeds byte allowance", budget!, "physical-bytes");
privateBytes += metadata.size;
if (!Number.isSafeInteger(privateBytes)) throw new Error("Invalid private cell evidence byte accounting");
const asset = cellTraceAsset(relative, input.privateCell.owner, input.privateCell.layout);
if (asset.reserved) {
const bytes = await readPrivateCellTraceAsset({ ...input.privateCell, relative, file: source, remaining: metadata.size, secrets: input.secrets });
await mkdir(path.dirname(destination), { recursive: true, mode: 0o700 });
await writeFile(destination, bytes!, { mode: 0o600 }); files.push(relative); continue;
}
if (!isStandardEvidenceFile(relative)) throw new Error("Unsupported private cell evidence carrier");
}
await mkdir(path.dirname(destination), { recursive: true });
if (ACTIVE_CONTENT_EXTENSIONS.has(extension)) {
// SVG can execute script when opened directly from an artifact. Keep the
// source in the disposable private attempt directory, but never admit it
// to the sanitized CI artifact.
continue;
} else if (TEXT_EXTENSIONS.has(extension)) {
const raw = await readFile(source, "utf8");
if (input.privateCell && inspectPrivateFrameworkMetadata(relative.split(path.sep).join("/"), raw, input.secrets)) {
assertSecretFree(raw, input.secrets, "private framework metadata");
await copyFile(source, destination); files.push(relative); continue;
}
if (extension === ".html") {
const issue = await inspectEmbeddedReport(raw, input.secrets);
if (issue) { if (budget && budget.bound === "none") budget.bound = scannerBudgetBound(issue); leaks.push({ file: relative, reason: issue }); continue; }
}
const parsed = extension === ".json" ? JSON.parse(raw) : null;
const leak =
extension === ".json"
? findSecretLeakInJsonValues(parsed, input.secrets)
: findSecretLeak(raw, input.secrets);
if (leak) leaks.push({ file: relative, reason: leak });
if (registeredAttachments.has(relative.split(path.sep).join("/"))) {
if (extension === ".json" ? JSON.stringify(sanitizeJson(parsed, input.secrets)) !== JSON.stringify(parsed) : redactText(raw, input.secrets) !== raw) throw new Error("Private framework attachment requires redaction");
if (!leak) { await copyFile(source, destination); files.push(relative); }
continue;
}
// Redacting an already serialized JSON string can change escape
// boundaries around shell commands. Sanitize parsed values instead so
// uploaded snapshots stay valid JSON.
const safe =
extension === ".json"
? `${JSON.stringify(sanitizeJson(parsed, input.secrets), null, 2)}\n`
: redactText(raw, input.secrets);
if (extension === ".json") {
const safeLeak = findSecretLeakInJsonValues(
JSON.parse(safe),
input.secrets,
);
if (safeLeak)
throw new Error(`Secret leak in ${relative}: ${safeLeak}`);
} else {
assertSecretFree(safe, input.secrets, relative);
}
await writeFile(destination, safe, "utf8");
files.push(relative);
} else if (extension === ".zip") {
const leak = await inspectZip(source, input.secrets);
if (leak) {
if (budget && budget.bound === "none") budget.bound = scannerBudgetBound(leak);
leaks.push({ file: relative, reason: leak });
continue;
}
await copyFile(source, destination);
files.push(relative);
} else if (BINARY_EXTENSIONS.has(extension)) {
// This raw-byte scan catches embedded plaintext credentials, but cannot
// inspect rendered pixels. Provider/UI raster and video files remain in
// the access-controlled CI artifact. The public publisher creates its
// own synthetic summary image from fixed labels and numeric/status data.
const raw = await readFile(source);
const leak = findSecretLeak(raw, input.secrets);
if (leak) {
leaks.push({ file: relative, reason: leak });
continue;
}
await copyFile(source, destination);
files.push(relative);
}
}
if (budget) { budget.inventoryComplete = true; budget.scanFinished = true; }
const missing = input.expectPassScreenshot
? [
...REQUIRED_PASS_FILES.filter((required) => !files.includes(required) && !(compact && required === path.join("html-report", "index.html"))),
...(compact || files.some(
(file) =>
file.startsWith(`blob-report${path.sep}`) && file.endsWith(".zip"),
)
? []
: [path.join("blob-report", "*.zip")]),
]
: [];
const manifest = {
schema: "paperclip.runner-e2e.evidence/v1",
files: [...files].sort(),
leaks,
missing,
};
const manifestText = `${JSON.stringify(sanitizeJson(manifest, input.secrets), null, 2)}\n`;
const manifestLeak = findSecretLeakInJsonValues(
JSON.parse(manifestText),
input.secrets,
);
if (manifestLeak)
throw new Error(`Secret leak in evidence-manifest.json: ${manifestLeak}`);
await writeFile(
path.join(input.uploadDir, "evidence-manifest.json"),
manifestText,
"utf8",
);
files.push("evidence-manifest.json");
if (budget && input.expectPassScreenshot) budget.requiredEvidenceComplete = missing.length === 0 && leaks.length === 0;
return { files, leaks, missing, ...(budget ? { budget } : {}) };
}
export async function publishRunnerE2EAttemptEvidence(input: {
result: RunnerE2EResult;
privateDir: string;
uploadDir: string;
resultsRoot: string;
secrets: readonly string[];
}): Promise<RunnerE2EResult> {
let result = input.result;
const resultPath = path.join(input.privateDir, "result.json");
const writeResult = () => writeFile(resultPath,
`${JSON.stringify(sanitizeJson(result, input.secrets), null, 2)}\n`, "utf8");
await assertRunnerE2EResultsPath(input.resultsRoot, input.uploadDir);
await writeResult();
const evidence = await packageEvidence({ ...input, expectPassScreenshot: result.status === "passed" });
if (evidence.leaks.length > 0 || evidence.missing.length > 0) {
result = {
...result,
status: "failed",
failureClass: evidence.leaks.length > 0 ? "secret_leak" : "permanent_infrastructure",
error: evidence.leaks.length > 0
? `Secret leak rejected from evidence: ${evidence.leaks.map(leak => leak.file).join(", ")}`
: `Required evidence missing: ${evidence.missing.join(", ")}`,
};
await writeResult();
await assertRunnerE2EResultsPath(input.resultsRoot, input.uploadDir);
await packageEvidence({ ...input, expectPassScreenshot: false });
}
return result;
}