Files
PaperClipAI/tests/runner-e2e/chat-catalog-domains/privileged-docker-database.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

94 lines
5.9 KiB
TypeScript

import { createHash } from "node:crypto";
import { lstat, readFile, realpath } from "node:fs/promises";
import path from "node:path";
import {
assertOwnedContainer, assertOwnedVolume, currentDockerEngine, readDockerDatabaseReceipt,
runDocker, DockerCommandError, publicDockerDatabaseIdentity, type DockerDatabaseReceipt, type DockerRun,
} from "../docker-postgres.js";
const digest = (bytes: string) => createHash("sha256").update(bytes).digest("hex");
export function ownedClusterMatches(expected: DockerDatabaseReceipt["cluster"], actual: unknown) {
const value = actual as DockerDatabaseReceipt["cluster"];
return !!expected && !!value && /^18\.6(?:\s|$)/.test(value.version)
&& /^[0-9]+$/.test(value.systemIdentifier) && value.systemIdentifier === expected.systemIdentifier
&& value.version === expected.version && value.username === expected.username && value.database === expected.database;
}
export function ownedRestoreTarget(name: string, sourceDatabase: string, nonce: string) {
return name === `catalog_restore_${digest(nonce).slice(0, 24)}` && name !== sourceDatabase;
}
/** Finite operator control of the launcher's existing cluster. No Docker client,
* receipt, password or control callback is passed to the native agent. */
export async function prepareOwnedDockerDatabase(root: string, configFile: string, run: DockerRun = runDocker) {
const filename = path.join(root, "docker-postgres-private", "receipt.json");
const initial = await readDockerDatabaseReceipt(filename, root, run);
if (!initial.artifactVerified || !initial.cluster) throw new Error("Owned Docker artifact/cluster qualification is missing");
const connection = initial.connectionString!;
const configBytes = await readFile(configFile, "utf8");
const config = JSON.parse(configBytes);
if (config.database?.mode !== "postgres" || config.database.connectionString !== connection)
throw new Error("Current database config does not select the exact owned Docker cluster");
const receiptDigest = digest(await readFile(filename, "utf8"));
const configDigest = digest(configBytes);
async function privateFile(file: string, exactMode?: number) {
const entry = await lstat(file);
if (await realpath(file) !== file || !entry.isFile() || entry.isSymbolicLink() || entry.uid !== process.getuid?.()
|| exactMode !== undefined && (entry.mode & 0o777) !== exactMode)
throw new Error("Owned database private file identity changed");
}
async function verify(running?: boolean) {
const rootEntry = await lstat(root), directory = path.dirname(filename), dirEntry = await lstat(directory);
if (await realpath(root) !== root || !rootEntry.isDirectory() || rootEntry.uid !== process.getuid?.() || rootEntry.mode & 0o077
|| !path.basename(root).startsWith("paperclip-runner-e2e-") || !configFile.startsWith(root + path.sep)
|| await realpath(directory) !== directory || !dirEntry.isDirectory() || dirEntry.uid !== process.getuid?.() || dirEntry.mode & 0o077)
throw new Error("Owned Docker attempt root changed");
await privateFile(filename, 0o600); await privateFile(configFile);
const passwordFile = path.join(directory, "password"); await privateFile(passwordFile, 0o600);
if (digest(await readFile(filename, "utf8")) !== receiptDigest || digest(await readFile(configFile, "utf8")) !== configDigest
|| await readFile(passwordFile, "utf8") !== new URL(connection).password)
throw new Error("Owned Docker receipt/config/password changed");
const engine = await currentDockerEngine(run, { DOCKER_CONTEXT: initial.engine.context });
if (JSON.stringify(engine) !== JSON.stringify(initial.engine)) throw new Error("Owned Docker engine changed");
const container = JSON.parse(await run(["--context", initial.engine.context, "inspect", initial.containerId!]))[0];
assertOwnedContainer(container, initial, running);
assertOwnedVolume(JSON.parse(await run(["--context", initial.engine.context, "volume", "inspect", initial.volumeName]))[0], initial);
return container.State.Running === true;
}
async function verifyCluster() {
await verify(true);
const actual = JSON.parse(await run(["--context", initial.engine.context, "exec", initial.containerId!, "psql", "-X", "-U", initial.username,
"-d", initial.database, "-At", "-c", "SELECT json_build_object('version',current_setting('server_version'),'database',current_database(),'username',current_user,'systemIdentifier',(SELECT system_identifier::text FROM pg_control_system()))"]));
if (!ownedClusterMatches(initial.cluster, actual)) throw new Error("Owned PostgreSQL cluster identity changed");
}
await verifyCluster();
let stopRequested = false;
return {
connection, database: initial.database, verify: verifyCluster,
proof: { ...publicDockerDatabaseIdentity(initial), ownerMatches: true, directoryWithinOwnedRoot: true },
async stop() {
await verifyCluster(); stopRequested = true;
await run(["--context", initial.engine.context, "stop", "--time", "10", initial.containerId!], 30_000);
await verify(false);
},
async recover() {
if (!stopRequested) return;
if (!await verify()) {
// Revalidate immediately before the only supported state mutation.
await verify(false);
await run(["--context", initial.engine.context, "start", initial.containerId!], 20_000);
}
// pg_isready is bounded and does not imply cluster identity by itself.
const deadline = Date.now() + 30_000;
while (true) {
try { await verifyCluster(); break; }
catch (error) {
if (!(error instanceof DockerCommandError) || Date.now() >= deadline) throw new Error("Exact owned Docker database recovery is unconfirmed; retain private supervisor evidence");
await verify(true); // Changed custody is never retried as startup latency.
await new Promise(resolve => setTimeout(resolve, 250));
}
}
stopRequested = false;
},
};
}