Files
PaperClipAI/scripts/verify-grok-npm-install.mjs
T
Devin FoleyandPaperclip b7f41ee51d perf(ci): warm the public-install npm cache during pnpm build (#16204)
## Thinking Path

> - Paperclip is the open source app people use to manage AI agents for
work.
> - Each pull request runs the Trusted PR CI. The `ci / Build` lane is
the longest check, or close to the longest, in each green run.
> - On 2026-10-13, `ci / Build` took 415-446s in runs 38426120144,
38432086560 and 38423713537. Of that, the public-install verifier took
226-241s.
> - In the verifier, the consumer `npm install` took 119-129s. It
downloads the full public registry graph into an empty npm cache.
> - Source manifests set that graph. Thus CI can download it before
`pnpm build` makes any tarball.
> - This pull request fills the npm cache in the background while the
Rust restore, Evalbook and `pnpm build` run. Then the verifier reads
that cache.
> - The benefit is a shorter `ci / Build` lane. The verifier keeps all
its checks.

## Linked Issues or Issue Description

Refs #15213, #15849, #15868, #15888 (earlier changes that made this
verifier faster).

**Enhancement description**
The `ci / Build` lane runs the public-install verifier after `pnpm
build`. The verifier starts with an empty npm cache. Its consumer
install then downloads the full public registry graph: 119-129s of a
226-241s step.

**Proposed change**
Download the same registry graph into an npm cache in the background,
during the earlier part of the lane. Give that cache to the verifier.

**Why it helps**
The download moves off the critical path of the lane. A local test of
the same 628-package install took 51s with a cold cache and 21s with a
warm cache.

## What Changed

- Add `scripts/public-install-closure.mjs`. It holds the one definition
of the public package closure: `@paperclipai/server`, `paperclipai`, and
each public workspace package that they reach through `workspace:`
specs. It also gives the registry dependencies of that closure.
- Add `scripts/warm-public-install-npm-cache.mjs`. It installs those
registry dependencies into a temporary consumer, with scripts disabled,
to fill `PAPERCLIP_PUBLIC_INSTALL_NPM_CACHE`. Then it removes the
temporary consumer.
- `scripts/verify-grok-npm-install.mjs` now gets its closure from the
shared module. When `PAPERCLIP_PUBLIC_INSTALL_NPM_CACHE` is set, it uses
that directory as the consumer npm cache. When the variable is not set,
it uses a new empty cache, as before. `docker-runner-check.yml` and
`release-smoke.yml` do not change.
- `pr-trusted.yml` (Build job): a new step starts the warm-up in the
background after `Install dependencies`. The verify step waits for the
warm-up for a maximum of 180s and then gives the cache to the verifier.
If the warm-up fails, the step writes a notice and the verifier
continues as before.
- Add `scripts/__tests__/public-install-closure.test.mjs` and add it to
`test:release-registry`. Update the closure assertions in
`grok-public-install-sandbox.test.mjs`.

## Verification

- `node --test scripts/__tests__/public-install-closure.test.mjs`: 4/4
pass.
- `node --test scripts/__tests__/grok-public-install-sandbox.test.mjs`:
24/24 pass.
- `node --test scripts/__tests__/run-vitest-stable-shard.test.mjs
scripts/__tests__/e2e-shard.test.mjs
scripts/__tests__/runner-release-crate-cache.test.mjs`: all pass.
- `release-verify-workflow.test.mjs`: the two "publisher lockfile" tests
fail in my sandbox on clean master too, because they need pnpm. Two
canary git-fixture tests timed out once under load. They pass when they
run alone.
- I ran the warm-up against current master: 62 direct dependencies, 628
packages, 51s cold. The same install with the warm cache took 21s.
- Important: `pr.yml` calls `pr-trusted.yml@master`. Thus the CI of this
pull request runs the refactored verifier on its default path (empty
cache), but not the new warm-up steps. The warm-up steps start after
merge. To measure them, compare the `consumer_install_ms` value in the
`paperclip.public-npm-install.timing.v1` line of a Build job before and
after merge.

- Final review: all latest-head CI checks pass (54 successes, one
neutral result, two intentional skips). Fresh Apex is 5/5 on
`64d091b8b0c5359fa3dc0edb1fd7ef9d24294477` with no unresolved findings.
The two relevant local script suites pass all 28 tests. Current master
combines cleanly.

## Risks

- Low risk. The verifier result does not change. npm checks each cached
tarball against the integrity of its own resolution, and the consumer
directory and lock start empty.
- npm can use a cached registry packument that is a maximum of about 5
minutes old (registry `max-age`). This has the same effect as starting
the verifier a few minutes earlier.
- The warm-up runs at the same time as `pnpm build`. It uses network,
disk and some CPU. In the measured runs, `pnpm build` took 76-96s.
- If the warm-up is slow or fails, the verify step waits a maximum of
180s and then continues with what is in the cache.
- The verifier receipt still reports `cleanNpmInstall: true`. The
consumer install is clean. Only the download cache is shared.

## Expected time savings

- `ci / Build`: an estimated 65-85s less, from 415-446s to about
340-370s. The consumer install goes from 119-129s to about 40-55s, from
the local cold/warm ratio.
- Full PR time: after this change, `ci / Build` is no longer the longest
lane. The longest lanes become `General tests (server N/12)` and `Verify
Paperclip Runner (vitest 1/2)` at about 405-496s. Thus the full PR time
decreases only when Build is the critical path. That is approximately
0-30s in the runs above.

## Model Used

- Anthropic Claude Opus 5.5 (`claude-opus-5-5`), with tool use and code
execution in a Paperclip agent run.

OpenAI GPT-6 (`gpt-6`) through Codex assisted the final repository
review and targeted script validation. This session does not expose an
exact context-window size.

## 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
- [x] All Paperclip CI gates are green
- [x] 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: Paperclip <noreply@paperclip.ing>
2026-10-13 17:18:44 -07:00

454 lines
35 KiB
JavaScript

#!/usr/bin/env node
// Run on a disposable Linux verification host after pnpm build. No publication,
// credentials, inference, or changes to release versions. Native provisioning is
// explicit and separate from npm installation, and is removed in finally.
import assert from 'node:assert/strict';
import { execFileSync } from 'node:child_process';
import { chmodSync, cpSync, existsSync, mkdirSync, mkdtempSync, readFileSync, readdirSync, realpathSync, renameSync, rmSync, symlinkSync, writeFileSync } from 'node:fs';
import { createHash } from 'node:crypto';
import { createServer } from 'node:net';
import { tmpdir } from 'node:os';
import { stripTypeScriptTypes } from 'node:module';
import { basename, dirname, join, resolve } from 'node:path';
import { fileURLToPath } from 'node:url';
import { packConcurrency, packStagedPackages, packTimings } from './pack-staged-packages.mjs';
import { materializePublishManifest } from './prepare-bundled-package.mjs';
import { publicInstallClosure } from './public-install-closure.mjs';
import { GROK_PUBLIC_INSTALL_IMAGE, GROK_PUBLIC_INSTALL_LIFECYCLE, assertInstalledNativeConsumerContinuity, assertNoBundledCodexPayloads, grokConsumerDockerArgs, installedCodexProbeSource, installedDaemonProbeSource, macNodeHeadersDirectory, macPublicInstallLifecycleProfile, macInstalledStartupProbeSource, runMacInitialInstallWithDiagnostics } from './grok-public-install-sandbox.mjs';
import { startPrefetch, stopPrefetches } from './prefetch-child.mjs';
import { retainRunnerQualificationPackages } from './retain-runner-qualification-packages.mjs';
import { bundledInstallTimings, startBundledInstalls } from './stage-bundled-packages.mjs';
const repo = resolve(dirname(fileURLToPath(import.meta.url)), '..');
async function verifyMacConsumer(packageDirectory, sourceRevision, receiptPath) {
assert.equal(process.env.GITHUB_ACTIONS, 'true', 'Mac clean consumers run only on disposable hosted CI');
assert.equal(process.platform, 'darwin');
assert.ok(['arm64', 'x64'].includes(process.arch));
assert.match(sourceRevision, /^[a-f0-9]{40}$/);
const packages = realpathSync(packageDirectory);
const manifest = JSON.parse(readFileSync(join(packages, 'manifest.json'), 'utf8'));
assert.equal(manifest.schema, 'paperclip.runner.public-packages/v1');
assert.equal(manifest.sourceRevision, sourceRevision);
assert.ok(manifest.archives.some(entry => entry.name === '@paperclipai/server'));
assert.ok(manifest.archives.some(entry => entry.name === 'paperclipai'));
assert.equal(new Set(manifest.archives.map(entry => entry.name)).size, manifest.archives.length);
const root = realpathSync(mkdtempSync(join(tmpdir(), 'paperclip-mac-public-install-')));
try {
const consumer = join(root, 'consumer'), home = join(root, 'home'), tools = join(root, 'tools');
for (const directory of [consumer, home, tools, join(root, 'tmp'), join(root, 'cache')]) mkdirSync(directory);
const npmCli = realpathSync(execFileSync('/usr/bin/which', ['npm'], { encoding: 'utf8' }).trim());
symlinkSync(process.execPath, join(tools, 'node'));
const env = { PATH: `${tools}:/usr/bin:/bin:/usr/sbin:/sbin`, HOME: home, TMPDIR: join(root, 'tmp'),
npm_config_cache: join(root, 'cache'), npm_config_audit: 'false', npm_config_fund: 'false' };
const run = (command, args, cwd = consumer, runEnv = env) => execFileSync(command, args, { cwd, env: runEnv,
timeout: 10 * 60_000, maxBuffer: 32 * 1024 * 1024, encoding: 'utf8' });
const tarballs = manifest.archives.map(entry => {
assert.match(entry.file, /^[A-Za-z0-9._-]+\.tgz$/); assert.match(entry.sha256, /^[a-f0-9]{64}$/);
const path = join(packages, entry.file);
assert.equal(realpathSync(path), path);
assert.equal(createHash('sha256').update(readFileSync(path)).digest('hex'), entry.sha256);
assertNoBundledCodexPayloads(run('/usr/bin/tar', ['-tzf', path]).trim().split('\n'));
return path;
});
writeFileSync(join(consumer, 'package.json'), JSON.stringify({ private: true, type: 'module' }));
const sentinelSource = join(root, 'lifecycle-sentinel'); mkdirSync(sentinelSource);
writeFileSync(join(sentinelSource, 'package.json'), JSON.stringify({ name: 'paperclip-verification-lifecycle-sentinel', version: '1.0.0',
scripts: { postinstall: 'node -e "require(\'node:fs\').writeFileSync(\'lifecycle-ran\', \'ok\')"' } }));
run(process.execPath, [npmCli, 'pack', '--ignore-scripts', '--pack-destination', root], sentinelSource);
const sentinelTar = join(root, readdirSync(root).find(name => name.endsWith('.tgz')));
runMacInitialInstallWithDiagnostics(() => run(process.execPath, [npmCli, 'install', '--ignore-scripts', '--omit=dev', '--timing', ...tarballs, sentinelTar]),
{ sourceRevision, platform: process.platform, architecture: process.arch, cacheDirectory: join(root, 'cache'), receiptPath });
const sentinel = join(consumer, 'node_modules/paperclip-verification-lifecycle-sentinel/lifecycle-ran');
assert.equal(existsSync(sentinel), false);
const lock = readFileSync(join(consumer, 'package-lock.json'), 'utf8');
const profile = join(root, 'lifecycle.sb'); writeFileSync(profile, macPublicInstallLifecycleProfile(root));
const listener = createServer(socket => socket.end());
await new Promise((resolve, reject) => { listener.once('error', reject); listener.listen(0, '127.0.0.1', resolve); });
try {
const port = listener.address().port;
const networkProbe = `import assert from 'node:assert/strict'; import net from 'node:net';
const socket = net.connect({host:'127.0.0.1',port:${port}}); socket.setTimeout(3000,()=>process.exit(2));
socket.once('connect',()=>{socket.destroy(); ${"process.argv[1] === 'denied' ? process.exit(3) : process.exit(0)"};});
socket.once('error',error=>{assert.equal(process.argv[1],'denied'); assert.equal(error.code,'EPERM');});`;
run(process.execPath, ['--input-type=module', '-e', networkProbe, 'allowed']);
run('/usr/bin/sandbox-exec', ['-f', profile, process.execPath, '--input-type=module', '-e', networkProbe, 'denied']);
} finally { await new Promise(resolve => listener.close(resolve)); }
const nodeHeaders = macNodeHeadersDirectory(process.execPath, process.versions.node);
run('/usr/bin/sandbox-exec', ['-f', profile, process.execPath, npmCli, 'rebuild', '--offline', '--ignore-scripts=false', '--dangerously-allow-all-scripts', '--foreground-scripts'],
consumer, { ...env, npm_config_nodedir: nodeHeaders });
assert.equal(readFileSync(sentinel, 'utf8'), 'ok');
assert.equal(readFileSync(join(consumer, 'package-lock.json'), 'utf8'), lock);
for (const entry of manifest.archives) {
assert.equal(JSON.parse(readFileSync(join(consumer, 'node_modules', entry.name, 'package.json'))).version, manifest.releaseVersion);
}
const server = join(consumer, 'node_modules/@paperclipai/server'), installed = join(server, 'dist/vendor/paperclip-runner');
const probe = (name, source) => { const file = join(root, name); writeFileSync(file, source); return JSON.parse(run(process.execPath, [file]).trim()); };
const daemon = probe('daemon-probe.mjs', installedDaemonProbeSource(installed, consumer, sourceRevision));
const codex = probe('codex-probe.mjs', installedCodexProbeSource(join(installed, 'index.js'), consumer, undefined, { nativeCodex: true }));
const startupScript = macInstalledStartupProbeSource(consumer, server, daemon.sha256, join(repo, 'tests/runner-e2e/installed-cli.ts'));
const startupFile = join(root, 'startup-probe.mts'); writeFileSync(startupFile, startupScript);
const startup = JSON.parse(run(process.execPath, ['--import', 'tsx', startupFile], join(repo, 'server')).trim());
const receipt = { schema: 'paperclip.runner.mac-public-install.v1', sourceRevision, platform: process.platform, architecture: process.arch,
cleanNpmInstall: true, packageCount: manifest.archives.length, codexPackedPayloadsOmitted: true,
consumerLockPreserved: true, lifecycleSentinelVerified: true, lifecycleNodeHeadersVerified: true, lifecycleNodeVersion: process.versions.node,
lifecycleNetwork: 'sandbox-exec denied (loopback negative control)',
...daemon, ...codex, startup, providerCalls: 0 };
writeFileSync(receiptPath, JSON.stringify(receipt, null, 2) + '\n');
console.log(JSON.stringify(receipt));
} finally { rmSync(root, { recursive: true, force: true }); }
}
async function verifyLinuxArmConsumer(packageDirectory, sourceRevision, receiptPath) {
assert.equal(process.env.GITHUB_ACTIONS, 'true', 'Native clean consumers run only on disposable hosted CI');
assert.equal(process.platform + '-' + process.arch, 'linux-arm64', 'ARM qualification requires a native ARM host');
assert.equal(process.versions.node.split('.')[0], '24', 'ARM host startup requires the qualified Node24 runtime');
assert.match(sourceRevision, /^[a-f0-9]{40}$/);
const packages = realpathSync(packageDirectory);
const manifest = JSON.parse(readFileSync(join(packages, 'manifest.json'), 'utf8'));
assert.equal(manifest.schema, 'paperclip.runner.public-packages/v1');
assert.equal(manifest.sourceRevision, sourceRevision);
assert.ok(manifest.archives.some(entry => entry.name === '@paperclipai/server'));
assert.ok(manifest.archives.some(entry => entry.name === 'paperclipai'));
assert.equal(new Set(manifest.archives.map(entry => entry.name)).size, manifest.archives.length);
const root = realpathSync(mkdtempSync(join(tmpdir(), 'paperclip-arm-public-install-')));
try {
const assets = join(root, 'assets'), consumer = join(root, 'consumer'), cache = join(root, 'cache');
for (const directory of [assets, consumer, cache]) mkdirSync(directory);
const run = (command, args, cwd = root) => execFileSync(command, args, { cwd,
env: { PATH: process.env.PATH, HOME: root, TMPDIR: root }, timeout: 10 * 60_000,
maxBuffer: 32 * 1024 * 1024, encoding: 'utf8' });
const platform = 'linux/arm64';
run('docker', ['pull', '--platform', platform, GROK_PUBLIC_INSTALL_IMAGE]);
const isolated = (command, options = {}) => run('docker', grokConsumerDockerArgs({ assets, consumer, cache,
uid: process.getuid(), gid: process.getgid(), platform, command, ...options }));
for (const entry of manifest.archives) {
assert.match(entry.file, /^[A-Za-z0-9._-]+\.tgz$/); assert.match(entry.sha256, /^[a-f0-9]{64}$/);
const tarball = join(packages, entry.file);
assert.equal(realpathSync(tarball), tarball);
assert.equal(createHash('sha256').update(readFileSync(tarball)).digest('hex'), entry.sha256);
assertNoBundledCodexPayloads(run('tar', ['-tzf', tarball]).trim().split('\n'));
cpSync(tarball, join(assets, entry.file)); chmodSync(join(assets, entry.file), 0o644);
}
writeFileSync(join(consumer, 'package.json'), JSON.stringify({ private: true, type: 'module' }));
const sentinelSource = join(root, 'lifecycle-sentinel'); mkdirSync(sentinelSource);
writeFileSync(join(sentinelSource, 'package.json'), JSON.stringify({ name: 'paperclip-verification-lifecycle-sentinel', version: '1.0.0',
scripts: { postinstall: 'node -e "require(\'node:fs\').writeFileSync(\'lifecycle-ran\', \'ok\')"' } }));
run('npm', ['pack', '--ignore-scripts', '--pack-destination', assets], sentinelSource);
isolated(['npm', 'install', '--ignore-scripts', '--omit=dev', ...readdirSync(assets).filter(file => file.endsWith('.tgz')).map(file => `/packages/${file}`)], { download: true });
const sentinel = join(consumer, 'node_modules/paperclip-verification-lifecycle-sentinel/lifecycle-ran');
assert.equal(existsSync(sentinel), false);
const lock = readFileSync(join(consumer, 'package-lock.json'), 'utf8');
isolated(GROK_PUBLIC_INSTALL_LIFECYCLE);
assert.equal(readFileSync(sentinel, 'utf8'), 'ok');
assert.equal(readFileSync(join(consumer, 'package-lock.json'), 'utf8'), lock);
for (const entry of manifest.archives) {
assert.equal(JSON.parse(readFileSync(join(consumer, 'node_modules', entry.name, 'package.json'))).version, manifest.releaseVersion);
}
const server = '/consumer/node_modules/@paperclipai/server', installed = server + '/dist/vendor/paperclip-runner';
const probe = (name, source, options = {}) => { writeFileSync(join(assets, name), source); return JSON.parse(isolated(['node', '/packages/' + name], options).trim()); };
const daemon = probe('daemon-probe.mjs', installedDaemonProbeSource(installed, '/consumer', sourceRevision));
assert.equal(daemon.target, 'linux-arm64');
const codex = probe('codex-probe.mjs', installedCodexProbeSource(installed + '/index.js', '/consumer', undefined, { nativeCodex: true }));
cpSync(join(repo, 'tests/runner-e2e/installed-cli.ts'), join(assets, 'installed-cli.ts'));
writeFileSync(join(assets, 'types.js'), stripTypeScriptTypes(readFileSync(join(repo, 'tests/runner-e2e/types.ts'), 'utf8')));
const home = join(root, 'host-home'), hostTmp = join(root, 'host-tmp');
for (const directory of [home, hostTmp]) mkdirSync(directory, { mode: 0o700 });
const hostEnv = { PATH: `${dirname(process.execPath)}:/usr/bin:/bin:/usr/sbin:/sbin`, HOME: home, TMPDIR: hostTmp, LC_ALL: 'C', TZ: 'UTC' };
const hostProbe = (name, source) => {
const file = join(assets, name); writeFileSync(file, source);
return JSON.parse(execFileSync(process.execPath, [file], { cwd: consumer, env: hostEnv,
timeout: 120_000, maxBuffer: 1024 * 1024, encoding: 'utf8' }).trim());
};
const hostServer = join(consumer, 'node_modules/@paperclipai/server'), hostInstalled = join(hostServer, 'dist/vendor/paperclip-runner');
const hostDaemon = () => hostProbe('host-daemon-probe.mjs', installedDaemonProbeSource(hostInstalled, consumer, sourceRevision));
const hostCodex = () => hostProbe('host-codex-probe.mjs', installedCodexProbeSource(join(hostInstalled, 'index.js'), consumer, undefined, { nativeCodex: true }));
const digest = value => createHash('sha256').update(value).digest('hex');
const beforeCodex = hostCodex();
const before = { lockSha256: digest(lock), runnerSha256: hostDaemon().sha256.slice(7), codexCommandSha256: beforeCodex.codexCommandSha256, codexNativeExecutableSha256: beforeCodex.codexNativeExecutableSha256 };
assert.equal(before.runnerSha256, daemon.sha256.slice(7), 'Docker and host must select the same packaged daemon bytes');
assert.equal(before.codexCommandSha256, codex.codexCommandSha256, 'Docker and host must select the same official Codex command bytes');
assert.equal(before.codexNativeExecutableSha256, codex.codexNativeExecutableSha256, 'Docker and host must select the same official native Codex payload bytes');
const hostHeader = process.report.getReport().header;
assert.match(hostHeader.glibcVersionRuntime || '', /^\d+\.\d+$/, 'Native host startup must report its actual GNU libc ABI');
const startup = hostProbe('host-startup-probe.mjs', macInstalledStartupProbeSource(consumer, hostServer, daemon.sha256, join(assets, 'installed-cli.ts')));
const afterCodex = hostCodex();
const after = { lockSha256: digest(readFileSync(join(consumer, 'package-lock.json'))), runnerSha256: hostDaemon().sha256.slice(7), codexCommandSha256: afterCodex.codexCommandSha256, codexNativeExecutableSha256: afterCodex.codexNativeExecutableSha256 };
assertInstalledNativeConsumerContinuity(before, after);
assert.equal(startup.runnerSha256, before.runnerSha256);
assert.equal(startup.codexSha256, before.codexCommandSha256);
const receipt = { schema: 'paperclip.runner.linux-arm64-public-install.v1', sourceRevision, platform: process.platform, architecture: process.arch,
cleanNpmInstall: true, packageCount: manifest.archives.length, codexPackedPayloadsOmitted: true,
consumerImage: GROK_PUBLIC_INSTALL_IMAGE, consumerImagePlatform: platform,
consumerLockPreserved: true, lifecycleSentinelVerified: true, lifecycleNetwork: 'none',
startupEnvironment: 'native host, same installed graph, isolated empty homes',
startupHost: { nodeVersion: process.versions.node, nodeModulesAbi: process.versions.modules, glibcVersionRuntime: hostHeader.glibcVersionRuntime, kernelRelease: hostHeader.osRelease },
installedIdentityBefore: before, installedIdentityAfter: after,
...daemon, ...codex, startup, providerCalls: 0 };
writeFileSync(receiptPath, JSON.stringify(receipt, null, 2) + '\n');
console.log(JSON.stringify(receipt));
} finally { rmSync(root, { recursive: true, force: true }); }
}
if (process.argv[2] === '--linux-arm64-consumer') {
assert.equal(process.argv.length, 6, 'Usage: --linux-arm64-consumer PACKAGE_DIRECTORY SOURCE_SHA RECEIPT_PATH');
await verifyLinuxArmConsumer(process.argv[3], process.argv[4], process.argv[5]);
} else if (process.argv[2] === '--mac-consumer') {
assert.equal(process.argv.length, 6, 'Usage: --mac-consumer PACKAGE_DIRECTORY SOURCE_SHA RECEIPT_PATH');
await verifyMacConsumer(process.argv[3], process.argv[4], process.argv[5]);
} else {
assert.equal(process.platform, 'linux', 'Run this verification on disposable EC2 Linux, not a developer host');
const root = mkdtempSync(join(tmpdir(), 'paperclip-grok-public-install-'));
const prerequisite = join(root, 'native/grok');
const env = { ...process.env, NODE_PATH: '', PAPERCLIP_RELEASE_REUSE_UI_DIST: '1', npm_config_ignore_scripts: 'false', npm_config_audit: 'false', npm_config_fund: 'false' };
const run = (cmd, args, cwd = root) => execFileSync(cmd, args, { cwd, env, stdio: 'pipe', maxBuffer: 32 * 1024 * 1024 });
const sourceRevision = run('git', ['rev-parse', 'HEAD'], repo).toString().trim();
const releaseVersion = `0.0.0-grok-verify.${sourceRevision.slice(0, 12)}`;
// The two network downloads the probes need do not depend on anything packed
// below, so they overlap the CPU-bound staging instead of running after it.
// The consumer image is pinned by digest and the Grok binary is checked against
// its pinned sha256 by the provisioning script, so an early start changes no
// verified property. Neither touches the consumer, the assets, nor the
// `native/grok` path that the post-install assertion inspects.
const prefetched = join(root, 'prefetch'); mkdirSync(prefetched);
const stagedGrok = join(prefetched, 'grok');
// Every background child, downloads and bundled-dependency installs alike, is
// stopped in `finally`; `downloads` is the subset the timing line reports as
// prefetches.
const prefetches = [];
const downloads = [];
const timing = {};
// A prefetch failure surfaces where its result is awaited, never as an
// unhandled rejection that would mask an earlier staging error.
const prefetch = (name, cmd, args) => {
const handle = startPrefetch({ name, cmd, args, cwd: root, env, logPath: join(prefetched, `${name}.log`) });
prefetches.push(handle);
downloads.push(handle);
return handle.done;
};
let phaseStarted = performance.now();
const phase = name => { const now = performance.now(); timing[`${name}_ms`] = Math.round(now - phaseStarted); phaseStarted = now; };
const awaited = async (name, promise) => { const started = performance.now(); await promise; timing[`${name}_wait_ms`] = Math.round(performance.now() - started); };
try {
const imageReady = prefetch('image', 'docker', ['pull', '--platform', 'linux/amd64', GROK_PUBLIC_INSTALL_IMAGE]);
// Provision as the unprivileged verification user, never into the host's /opt.
// Only the positive probe sees this file at the canonical sandbox path.
const grokReady = prefetch('grok', process.execPath, [join(repo, 'packages/paperclip-runner/scripts/provision-grok.mjs'), stagedGrok]);
// The server, the CLI and every public workspace package they reach through
// `workspace:` specs. The npm cache warm-up reads the same closure, so both
// take it from one module.
const { packages, needed } = publicInstallClosure(repo);
// release.sh also builds the standalone public packages (the sandbox-provider
// plugins outside the pnpm workspace) before it stages anything, because it
// publishes them. Nothing staged here reaches them: `needed` is closed under
// `workspace:` specs, so every package staged below is a workspace package
// that `pnpm build` already built, and that build would only cost one
// network install and compile per plugin whose output no tarball reads.
// Canary Dry Run still runs it through release.sh. The one path that does
// read a standalone build is qualification-package retention at the end:
// it packs the Daytona plugin from its `dist`, which only this build
// produces, so the build runs when retention is requested and nowhere else.
const retainQualificationPackages = Boolean(process.env.PAPERCLIP_RUNNER_QUALIFICATION_PACKAGES_DIR);
if (retainQualificationPackages) {
const standaloneStarted = performance.now();
run(process.execPath, [join(repo, 'scripts/build-standalone-public-packages.mjs')], repo);
timing.stage_standalone_build_ms = Math.round(performance.now() - standaloneStarted);
}
// The server's `ui-dist` payload is prepared here; with the reuse flag set
// above, that copies the `ui/dist` that `pnpm build` produced instead of
// building the UI again.
const uiDistStarted = performance.now();
run('bash', [join(repo, 'scripts/prepare-server-ui-dist.sh')], repo);
timing.stage_ui_dist_ms = Math.round(performance.now() - uiDistStarted);
// Match release.sh's unified versioning in temporary staging directories.
// Source manifests remain untouched, including independently versioned SDKs.
// Each package's published files are staged into `source-<index>` and its
// pack input into `package-<index>`, with <index> in dependency order so the
// tarball list keeps matching `needed`. The packages that bundle dependencies
// each need a host-side `npm install` of their bundled closure; those installs
// write only their own `package-<index>` and read nothing another package
// produces, so they start first and run together while the other packages
// are copied, instead of one after another inside one loop. Packing is a
// Node start-up plus CPU-bound gzip per package that reads only the package's
// own directory, so the packs also run together, each into its own
// `pack-<index>`, and the tarballs are checked in order once all are made.
const entries = [...needed].map((name, index) => {
const { dir, manifest } = packages.get(name);
const bundled = (manifest.bundleDependencies ?? manifest.bundledDependencies ?? []).length > 0;
return { name, index, dir, manifest, bundled, target: join(root, `package-${index}`), stagedSource: join(root, `source-${index}`), destination: join(root, `pack-${index}`) };
});
const stageSource = ({ dir, manifest, stagedSource }) => {
mkdirSync(stagedSource);
for (const file of manifest.files ?? ['dist']) {
// release.sh stages runtime skills into these public packages before pack.
const releaseSkills = file === 'skills' && ['server', 'packages/adapters/claude-local', 'packages/adapters/codex-local'].includes(dir);
// Other adapters declare an optional skills directory that npm pack omits
// when absent. Do not fabricate extra release payloads for those adapters.
if (file === 'skills' && !releaseSkills && !existsSync(join(repo, dir, file))) continue;
cpSync(releaseSkills ? join(repo, 'skills') : join(repo, dir, file), join(stagedSource, file), { recursive: true });
}
const releaseManifest = { ...manifest, version: releaseVersion };
writeFileSync(join(stagedSource, 'package.json'), JSON.stringify(releaseManifest));
return releaseManifest;
};
const sourcesStarted = performance.now();
for (const entry of entries.filter(entry => entry.bundled)) stageSource(entry);
const installsStarted = performance.now();
const installs = startBundledInstalls(entries.filter(entry => entry.bundled), { logDir: prefetched, cwd: repo });
prefetches.push(...installs.map(({ handle }) => handle));
for (const entry of entries.filter(entry => !entry.bundled)) {
const releaseManifest = stageSource(entry);
mkdirSync(entry.target);
cpSync(entry.stagedSource, entry.target, { recursive: true });
writeFileSync(join(entry.target, 'package.json'), JSON.stringify(materializePublishManifest(releaseManifest)));
}
timing.stage_sources_ms = Math.round(performance.now() - sourcesStarted);
await awaited('bundled', Promise.all(installs.map(({ handle }) => handle.done)));
timing.stage_bundled_ms = Math.round(performance.now() - installsStarted);
const packStarted = performance.now();
// The bundled packages make the largest tarballs, so they are launched first
// and the short packs fill the remaining slots behind them. Each pack is
// registered for cleanup as it starts, so a failure stops the ones still
// running before the scratch root is removed. The results are put back into
// dependency order before the tarballs are checked and copied.
const concurrency = packConcurrency();
const packs = await packStagedPackages([...entries].sort((a, b) => Number(b.bundled) - Number(a.bundled)), { logDir: prefetched, env, concurrency, register: handle => prefetches.push(handle) });
packs.sort((a, b) => a.index - b.index);
const checksStarted = performance.now();
const tarballs = packs.map(({ tarball }) => tarball);
for (const tarball of tarballs) assertNoBundledCodexPayloads(run('tar', ['-tzf', tarball]).toString().trim().split('\n'));
timing.stage_pack_check_ms = Math.round(performance.now() - checksStarted);
timing.stage_pack_ms = Math.round(performance.now() - packStarted);
Object.assign(timing, packTimings(packs, concurrency));
if (process.env.PAPERCLIP_RUNNER_PUBLIC_PACKS_DIR) {
const output = resolve(process.env.PAPERCLIP_RUNNER_PUBLIC_PACKS_DIR);
assert.equal(JSON.parse(readFileSync(join(repo, 'server/dist/build-info.json'))).commit, sourceRevision);
assert.ok(!existsSync(output), 'Never replace retained public packages');
mkdirSync(output, { recursive: true });
const archives = [...needed].map((name, index) => {
const file = basename(tarballs[index]); cpSync(tarballs[index], join(output, file));
return { name, file, sha256: createHash('sha256').update(readFileSync(tarballs[index])).digest('hex') };
});
writeFileSync(join(output, 'manifest.json'), JSON.stringify({
schema: 'paperclip.runner.public-packages/v1', sourceRevision, releaseVersion,
publicInstallationVerified: false, providerCalls: 0, archives,
}, null, 2) + '\n');
}
const assets = join(root, 'assets'); mkdirSync(assets, { mode: 0o755 });
const consumer = join(root, 'consumer'); mkdirSync(consumer);
// CI can point this at a cache that scripts/warm-public-install-npm-cache.mjs
// filled with the same public registry graph while `pnpm build` ran. npm
// checks every cached tarball against the integrity its own resolution
// records, so a warm cache changes where bytes come from, never what is
// installed; the consumer directory and its lock still start empty.
const cache = process.env.PAPERCLIP_PUBLIC_INSTALL_NPM_CACHE ? resolve(process.env.PAPERCLIP_PUBLIC_INSTALL_NPM_CACHE) : join(root, 'cache');
mkdirSync(cache, { recursive: true });
writeFileSync(join(consumer, 'package.json'), JSON.stringify({ private: true, type: 'module' }));
for (const tarball of tarballs) {
const destination = join(assets, basename(tarball)); cpSync(tarball, destination); chmodSync(destination, 0o644);
}
// Prove lifecycle execution is real even if npm changes its script defaults.
const sentinelSource = join(root, 'lifecycle-sentinel'); mkdirSync(sentinelSource);
writeFileSync(join(sentinelSource, 'package.json'), JSON.stringify({
name: 'paperclip-verification-lifecycle-sentinel', version: '1.0.0', private: true,
scripts: { postinstall: 'node -e "require(\'node:fs\').writeFileSync(\'lifecycle-ran\', \'ok\')"' },
}));
run('npm', ['pack', '--ignore-scripts', '--pack-destination', assets], sentinelSource);
const sentinel = join(consumer, 'node_modules/paperclip-verification-lifecycle-sentinel/lifecycle-ran');
const consumerUid = process.getuid();
const isolated = (command, options = {}) => run('docker', grokConsumerDockerArgs({ assets, consumer, cache, uid: consumerUid, gid: process.getgid(), command, ...options }));
phase('stage_packages');
await awaited('image', imageReady);
// npm resolution is intentionally the public consumer graph, not the pnpm
// workspace graph. Freeze that result before any lifecycle script can run.
isolated(['npm', 'install', '--ignore-scripts', '--omit=dev', ...readdirSync(assets).filter(file => file.endsWith('.tgz')).map(file => `/packages/${file}`)], { download: true });
phase('consumer_install');
assert.equal(existsSync(sentinel), false, 'Dependency download must not run lifecycle scripts');
const consumerLock = readFileSync(join(consumer, 'package-lock.json'), 'utf8');
// npm ci rejects bundled optional platform dependencies absent from its own
// generated lock. Rebuild runs the deferred install hooks on the installed
// graph without re-resolving it; network isolation and lock checks still hold.
isolated(GROK_PUBLIC_INSTALL_LIFECYCLE);
assert.equal(readFileSync(sentinel, 'utf8'), 'ok', 'Offline lifecycle scripts must actually execute');
assert.equal(readFileSync(join(consumer, 'package-lock.json'), 'utf8'), consumerLock, 'Lifecycle execution must preserve the resolved consumer lock');
phase('lifecycle_rebuild');
for (const name of needed) {
const installedManifest = JSON.parse(readFileSync(join(consumer, 'node_modules', name, 'package.json'), 'utf8'));
assert.equal(installedManifest.version, releaseVersion, `Installed release version for ${name}`);
}
assert.equal(existsSync(prerequisite), false, 'npm must not provision Grok');
const server = join(consumer, 'node_modules/@paperclipai/server');
const installed = join(server, 'dist/vendor/paperclip-runner');
const codexVersion = JSON.parse(readFileSync(join(repo, 'packages/paperclip-runner/acpx-profiles.json'), 'utf8')).profiles.codex.agentRuntimeVersion;
assert.match(codexVersion, /^\d+\.\d+\.\d+$/);
writeFileSync(join(assets, 'codex-probe.mjs'), installedCodexProbeSource(
'/consumer/node_modules/@paperclipai/server/dist/vendor/paperclip-runner/index.js', '/consumer', codexVersion), { mode: 0o644 });
const codex = JSON.parse(isolated(['node', '/packages/codex-probe.mjs']).toString().trim());
phase('codex_probe');
assert.equal(codex.codexConsumerHostOnly, true);
assert.equal(codex.codexPlatformPackageSource, 'official npm');
assert.ok(existsSync(join(installed, 'providers/grok/launcher.cjs')));
assert.equal(existsSync(join(consumer, 'node_modules/@paperclipai/grok-acp')), false);
assert.equal(existsSync(join(installed, 'providers/grok/bin')), false);
// Use real installed compiled code and its actual npm dependency graph. A
// separate process prevents module resolution from borrowing this checkout.
const probe = `
import assert from 'node:assert/strict';
import { verifyQualifiedAcpxInstallation } from '/consumer/node_modules/@paperclipai/server/dist/vendor/paperclip-runner/drivers/acpx/installation-integrity.js';
import { resolveQualifiedAcpxProfile } from '/consumer/node_modules/@paperclipai/server/dist/vendor/paperclip-runner/drivers/acpx/qualified-profiles.js';
const profile = resolveQualifiedAcpxProfile('grok', 'grok-4.7');
const inspect = () => verifyQualifiedAcpxInstallation(profile, () => { throw new Error('Grok must not resolve an npm package'); });
if (process.argv[2] === 'missing') await assert.rejects(inspect, /prerequisite missing/);
else { const installation = await inspect(); assert.equal(installation.agentServerPackageJsonPath, null); assert.equal(installation.agentRuntimePackageJsonPath, null); await (await installation.openCommand()).close(); }
`;
writeFileSync(join(assets, 'probe.mjs'), probe, { mode: 0o644 });
isolated(['node', '/packages/probe.mjs', 'missing']);
// The binary was provisioned outside `native/grok` while the packages were
// staged, so the assertion above still proves npm put nothing there. It moves
// into place only now, so only the positive probe sees it.
await awaited('grok', grokReady);
mkdirSync(dirname(prerequisite), { recursive: true });
renameSync(stagedGrok, prerequisite);
isolated(['node', '/packages/probe.mjs', 'present'], { prerequisite });
phase('grok_probes');
console.log(JSON.stringify({ schema: 'paperclip.grok.public-npm-install.v1', sourceRevision, releaseVersion, lifecycleScriptsEnabled: true, lifecycleSentinelVerified: true, lifecycleNetwork: 'none', consumerImage: GROK_PUBLIC_INSTALL_IMAGE, consumerUid, consumerLockPreserved: true, cleanNpmInstall: true, packageCount: needed.size, codexPackedPayloadsOmitted: true, codexPackedPackagesChecked: tarballs.length, ...codex, builtinLauncherPresent: true, separateGrokPackage: false, npmProvisionedBinary: false, missingPrerequisiteRejected: true, provisionedBinaryVerified: true, commandLeaseVerified: true, providerCalls: 0 }));
// Exercise Pi's public CLI and installed server, never a private workspace
// package or binary override. Public dependency downloads are explicit and
// isolated; the actual admission probe runs without a network or credentials.
const piProbe = join(assets, 'pi-public-install-probe.mjs');
cpSync(join(repo, 'scripts/pi-public-install-probe.mjs'), piProbe); chmodSync(piProbe, 0o644);
// Pi assembles a bundled runtime in scratch space before atomic publication.
// Keep the existing sandbox and memory bound; only this download needs more
// temporary capacity than the smaller offline lifecycle probes.
const setup = isolated(['node', '/consumer/node_modules/paperclipai/dist/index.js', 'runtime', 'setup', 'pi'], { download: true, temporarySizeMiB: 2048 }).toString();
const receipt = JSON.parse(setup.trim());
phase('pi_setup');
assert.equal(receipt.status, 'installed_verified');
assert.equal(receipt.target, 'linux-x64');
assert.equal(readFileSync(join(consumer, 'package-lock.json'), 'utf8'), consumerLock, 'Pi setup must preserve the consumer dependency graph');
// Verified launch leases also materialize the runtime in private scratch.
// Docker defaults tmpfs to noexec. The offline admission probe must execute
// its verified private snapshot while keeping lifecycle/download scratch noexec.
console.log(isolated(['node', '/packages/pi-public-install-probe.mjs', '/consumer/node_modules/@paperclipai/server'], { temporarySizeMiB: 2048, temporaryExecutable: true }).toString().trim());
phase('pi_probe');
// Wall time per phase, so the slow part of this step is visible in the CI log
// without re-running it. `*_prefetch_ms` runs from a download's start to its
// last output, and `*_wait_ms` is how long the main flow waited for a
// prefetch that had not finished yet; zero means the overlap hid it entirely.
// `stage_ui_dist_ms`, `stage_sources_ms`, `stage_bundled_ms` and
// `stage_pack_ms` split the staging phase, and `stage_bundled_<package>_ms`
// is each concurrent install. `stage_standalone_build_ms` appears only when
// qualification-package retention made the standalone build necessary.
const downloaded = Object.fromEntries(downloads.map(({ name, durationMs }) => [`${name}_prefetch_ms`, durationMs]));
console.log(JSON.stringify({ schema: 'paperclip.public-npm-install.timing.v1', ...timing, ...downloaded, ...bundledInstallTimings(installs) }));
if (retainQualificationPackages) {
const retained = retainRunnerQualificationPackages({ repo, output: process.env.PAPERCLIP_RUNNER_QUALIFICATION_PACKAGES_DIR, sourceRevision, releaseVersion, env,
publicArchives: [...needed].map((name, index) => ({ name, file: tarballs[index] })),
});
console.log(JSON.stringify(retained));
}
} finally {
// An earlier failure must not leave a prefetch running, and scratch is
// removed only once both children have stopped writing into it.
await stopPrefetches(prefetches);
rmSync(root, { recursive: true, force: true });
}
}