Files
PaperClipAI/packages/paperclip-runner/test/acpx-codex-package-contract.test.mjs
Dotta af3023f1e3 fix(runner): repair paid provider startup paths (#12769)
## Thinking Path

> - Paperclip manages AI agents that perform work.
> - Paperclip Runner connects durable task runs to local provider
processes.
> - The full-stack paid matrix exposed failures after the runner
integrity repair.
> - Verified JavaScript entrypoints lost their relative module graph
when Linux executed them through descriptor paths.
> - Returned provider startup errors also remained pending and became
indeterminate after recovery.
> - Sparse Codex tool lifecycle events lost the `write_document`
identity before task transcript projection.
> - This pull request repairs those three boundaries and makes the
structured-question fixture deterministic.
> - The benefit is repeatable provider startup, exact failure replay,
and correct inline Plan placement.

## Linked Issues or Issue Description

Refs #12721 and #12700.

**What happened?**

The paid runner matrix failed ACPX and OpenCode startup before provider
session creation. The runner journal then replaced the original startup
error with an indeterminate recovery result. Native Codex saved a Plan
but rendered it only as a fallback card. A legacy Claude waiting reply
could also echo the reserved terminal marker before the answer arrived.

**Expected behavior**

Verified JavaScript providers must start from immutable
descriptor-backed artifacts. Returned startup failures must persist as
terminal failed command results. Native tool lifecycle updates must
preserve the `write_document` boundary. Pre-answer fixture output must
not contain the reserved terminal marker.

**Steps to reproduce**

1. Run the local provider cells in the Runner Full-Stack E2E workflow.
2. Observe ACPX and OpenCode fail during `session.open` before provider
execution.
3. Observe recovery report `execution_indeterminate` instead of the
original startup error.
4. Run the native Codex Plan cell and observe the fallback Plan card
after the tool activity row.
5. Run the legacy Claude structured-question resume cell and observe an
early marker echo in waiting prose.

**Paperclip version or commit**

`0f9452101740835ce0b1488a204bf48acd5bafc3`

**Deployment mode**

Local development with the paid GitHub Actions acceptance workflow.

## What Changed

- Bundle the ACPX sidecar and OpenCode proxy as self-contained Node ESM
entrypoints before hashing and verified descriptor launch.
- Anchor ACPX dynamic provider package resolution at a
controller-derived provider-pack root and keep that root out of the
provider child environment.
- Persist executor-returned startup errors as redacted durable failed
command results while retaining indeterminate recovery for true process
death.
- Coalesce sparse native tool items by stable ID so a late
`write_document` name, input, and result reach the transcript boundary
once.
- Forbid the structured-question fixture from spelling or announcing its
reserved terminal marker before the user answers.

## Verification

- Rust and TypeScript regression tests cover durable failed replay, true
crash ambiguity, bundle closure, package-root derivation, environment
filtering, exact Codex tool lifecycle coalescing, and prompt
determinism.
- Local execution is intentionally limited to formatters and static diff
checks. GitHub Actions will run tests, type checks, builds, and security
checks.
- After ordinary CI is green, scoped paid cells will validate one ACPX
launch, one OpenCode launch, native Codex Plan projection, and legacy
Claude structured resume before a complete matrix rerun.
- Prior failing matrix:
https://github.com/paperclipai/paperclip/actions/runs/33682434315

## Risks

- Bundling changes the bytes covered by provider launch hashes.
Provider-pack generation already hashes the final built files.
- ACPX still loads qualified provider packages dynamically. The
controller supplies a normalized package root, while existing version,
digest, path, and descriptor checks remain active.
- Durable `failed` is terminal. Replays return the same redacted result
and do not execute the provider effect twice.

> 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

OpenAI Codex based on GPT-5 with agentic reasoning, repository
inspection, code editing, Git, parallel subagents, and GitHub Actions
coordination. The exact deployed snapshot and context-window size are
not exposed to this task.

## 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 linked related public work or described the bug in
this PR
- [x] I have not referenced internal or instance-local Paperclip issues
or links
- [x] My branch name describes the change and contains no internal
ticket id
- [ ] I have run tests locally and they pass (intentionally deferred to
GitHub Actions)
- [x] I have added or updated tests where applicable
- [x] No documentation change is required for this runtime repair
- [x] I have considered and documented the 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
2026-09-04 07:58:44 -05:00

204 lines
6.1 KiB
JavaScript

import assert from "node:assert/strict";
import { readFile } from "node:fs/promises";
import test from "node:test";
const runnerPackage = JSON.parse(
await readFile(new URL("../package.json", import.meta.url), "utf8"),
);
const rootPackage = JSON.parse(
await readFile(new URL("../../../package.json", import.meta.url), "utf8"),
);
const workspace = await readFile(
new URL("../../../pnpm-workspace.yaml", import.meta.url),
"utf8",
);
const acpxPatch = await readFile(
new URL("../../../patches/acpx@0.13.1.patch", import.meta.url),
"utf8",
);
const codexPatch = await readFile(
new URL(
"../../../patches/@agentclientprotocol__codex-acp@1.6.2.patch",
import.meta.url,
),
"utf8",
);
const claudePatch = await readFile(
new URL(
"../../../patches/@agentclientprotocol__claude-agent-acp@0.70.0.patch",
import.meta.url,
),
"utf8",
);
const qualifiedProfiles = await readFile(
new URL("../src/drivers/acpx/qualified-profiles.ts", import.meta.url),
"utf8",
);
const runnerdAcpxBackend = await readFile(
new URL(
"../runner/crates/runner-core/src/acpx_provider_backend.rs",
import.meta.url,
),
"utf8",
);
const providerPackBuilder = await readFile(
new URL("../scripts/build-provider-pack.mjs", import.meta.url),
"utf8",
);
const nativeSessionExecutor = await readFile(
new URL(
"../../../server/src/services/native-runtime/native-session-executor.ts",
import.meta.url,
),
"utf8",
);
test("the runner pins every qualified ACPX production dependency", () => {
assert.equal(runnerPackage.dependencies["@openai/codex"], undefined);
assert.equal(runnerPackage.optionalDependencies, undefined);
assert.equal(runnerPackage.dependencies.node, undefined);
assert.equal(runnerPackage.dependencies.acpx, "0.13.1");
assert.equal(
runnerPackage.dependencies["@agentclientprotocol/codex-acp"],
"1.6.2",
);
assert.equal(
runnerPackage.dependencies["@agentclientprotocol/claude-agent-acp"],
"0.70.0",
);
});
test("the patched Codex ACP command digest stays aligned across launch boundaries", () => {
const profileMatch =
/agent: "codex"[\s\S]*?commandDigest:\s*"(sha256:[a-f0-9]{64})"/.exec(
qualifiedProfiles,
);
assert.ok(profileMatch, "qualified Codex ACPX profile digest");
const digest = profileMatch[1];
assert.match(runnerdAcpxBackend, new RegExp(`"codex"[\\s\\S]*?${digest}`));
assert.match(
providerPackBuilder,
new RegExp(`acpxProfileDigests:[\\s\\S]*?codex:[\\s\\S]*?${digest}`),
);
assert.match(
nativeSessionExecutor,
new RegExp(
`REMOTE_PROVIDER_PACK_PROFILE_DIGESTS[\\s\\S]*?codex:[\\s\\S]*?${digest}`,
),
);
});
test("the package exposes only the reviewed runner CLI binaries", () => {
assert.deepEqual(runnerPackage.bin, {
"paperclip-runner-eval-session": "./dist/cli/eval-session.js",
"paperclip-runner-codex-proxy": "./dist/cli/codex-app-server-unix-proxy.js",
"paperclip-runner-acpx-sidecar": "./dist/cli/acpx-runtime-sidecar.js",
"paperclip-runner-opencode-proxy":
"./dist/cli/opencode-app-server-proxy.js",
});
});
test("old and new pnpm configuration both apply the exact runtime patches", () => {
assert.equal(
rootPackage.pnpm.patchedDependencies["acpx@0.13.1"],
"patches/acpx@0.13.1.patch",
);
assert.equal(
rootPackage.pnpm.patchedDependencies[
"@agentclientprotocol/claude-agent-acp@0.70.0"
],
"patches/@agentclientprotocol__claude-agent-acp@0.70.0.patch",
);
assert.equal(
rootPackage.pnpm.patchedDependencies[
"@agentclientprotocol/codex-acp@1.6.2"
],
"patches/@agentclientprotocol__codex-acp@1.6.2.patch",
);
assert.match(workspace, /acpx@0\.13\.1: patches\/acpx@0\.13\.1\.patch/);
assert.match(
workspace,
/codex-acp@1\.6\.2["']: patches\/@agentclientprotocol__codex-acp@1\.6\.2\.patch/,
);
assert.match(
workspace,
/claude-agent-acp@0\.70\.0["']: patches\/@agentclientprotocol__claude-agent-acp@0\.70\.0\.patch/,
);
assert.equal(rootPackage.pnpm.patchedDependencies["node@24.11.0"], undefined);
assert.doesNotMatch(workspace, /node@24\.11\.0:/);
assert.match(
providerPackBuilder,
/copyFileSync\(process\.execPath, stableNodeCommand\)/,
);
assert.match(codexPatch, /\+ "@openai\/codex": "0\.148\.0"/);
});
test("the ACPX patch preserves launch-only state and verified spawning", () => {
for (const token of [
"spawnEnvironment",
"spawnCwd",
"spawnAgent",
"SpawnOptionsWithoutStdio",
"this.options.spawnAgent",
]) {
assert.match(acpxPatch, new RegExp(token));
}
});
test("the ACPX patch fails closed on an invalid spawn environment", () => {
for (const token of [
"isPlainStringEnvironment",
"Object.getPrototypeOf(value)",
'Object.values(value).every((entry) => typeof entry === "string")',
"spawnEnvironment !== void 0",
"sourceEnvironment = spawnEnvironment()",
"ACPX spawn environment must be a plain record of string values",
]) {
assert.match(
acpxPatch,
new RegExp(token.replace(/[.*+?^${}()|[\]\\]/g, "\\$&")),
);
}
assert.doesNotMatch(acpxPatch, /spawnEnvironment\?\.\(\)/);
assert.doesNotMatch(
acpxPatch,
/spawnEnvironment \? \{ \.\.\.spawnEnvironment \} : \{ \.\.\.process\.env \}/,
);
});
test("the Codex patch enforces isolated instructions, tools, and skills", () => {
for (const token of [
"PAPERCLIP_ACPX_ISOLATED_CONTEXT",
"baseInstructions",
"rawInput: { serverName: params.serverName }",
'"features.apps": false',
"process.env.CODEX_HOME",
]) {
assert.match(
codexPatch,
new RegExp(token.replace(/[.*+?^${}()|[\]\\]/g, "\\$&")),
);
}
});
test("the Codex patch keeps MCP tool approvals on the governed permission channel", () => {
assert.match(
codexPatch,
/!context\.isToolApproval && this\.shouldUseAcpElicitation\(params\)/,
);
});
test("the Claude patch removes ambient project and local configuration", () => {
for (const token of [
"PAPERCLIP_ACPX_ISOLATED_CONTEXT",
'settingSources: ["user"]',
"userProvidedOptions?.mcpServers",
]) {
assert.match(
claudePatch,
new RegExp(token.replace(/[.*+?^${}()|[\]\\]/g, "\\$&")),
);
}
});