Files
PaperClipAI/tests/runner-e2e/chat-catalog-fingerprint.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

138 lines
11 KiB
TypeScript

import { CHAT_OPENCODE_MODEL } from "./chat-runtime-profile.js";
import { mkdtempSync, writeFileSync, rmSync, mkdirSync } from "node:fs";
import { tmpdir } from "node:os";
import path from "node:path";
import { pathToFileURL } from "node:url";
import { afterEach, expect, it, vi } from "vitest";
import { assertChatCatalogFingerprintLane, chatCatalogCaseFingerprint, chatCatalogSpawnedSourceEntries, fingerprintChatDependencyClosure, normalizedChatCatalogExecution } from "./chat-catalog-fingerprint.js";
import { findChatCatalogCase, loadChatCatalogImplementation, type ChatCatalogDescriptor } from "./chat-catalog-registry.js";
const directories: string[] = [];
afterEach(() => { vi.unstubAllEnvs(); for (const directory of directories.splice(0)) rmSync(directory, { recursive: true, force: true }); });
function fixture() {
const root = mkdtempSync(path.join(tmpdir(), "chat-fingerprint-")); directories.push(root);
const write = (file: string, body: string) => writeFileSync(path.join(root, file), body);
write("shared.ts", 'export { value } from "./transitive.js";');
write("selected.ts", 'import { value } from "./shared.js"; export const result = value;');
write("transitive.ts", "export const value = 1;");
write("unselected.ts", 'throw new Error("Unselected module must never execute");');
return { root, write, fingerprint: () => fingerprintChatDependencyClosure({ root, entries: ["selected.ts"], selected: { id: "selected", query: "Exact request" } }).digest };
}
it("invalidates shared, selected and transitive dependencies while ignoring an unloaded domain", () => {
for (const target of ["shared.ts", "selected.ts", "transitive.ts"]) {
const f = fixture(); const before = f.fingerprint();
f.write("unselected.ts", "export const changed = true;"); expect(f.fingerprint()).toBe(before);
f.write(target, "export const changed = true;"); expect(f.fingerprint()).not.toBe(before);
}
});
it("binds actual spawned supervisor, entry and sandbox bytes even when launch refers to them only by path", () => {
expect(chatCatalogSpawnedSourceEntries).toEqual(["tests/runner-e2e/server.ts", "tests/runner-e2e/server-entry.ts", "tests/runner-e2e/agent-filesystem-sandbox.py"]);
for (const changed of chatCatalogSpawnedSourceEntries) {
const f = fixture(); mkdirSync(path.join(f.root, "tests/runner-e2e"), { recursive: true });
for (const file of chatCatalogSpawnedSourceEntries) f.write(file, "export const behavior = 'original';");
const fingerprint = () => fingerprintChatDependencyClosure({ root: f.root, entries: ["selected.ts", ...chatCatalogSpawnedSourceEntries], selected: {} });
const before = fingerprint();
expect(before.sources.map(row => row.file)).toEqual(expect.arrayContaining([...chatCatalogSpawnedSourceEntries]));
f.write(changed, "export const behavior = 'changed executable';"); expect(fingerprint().digest).not.toBe(before.digest);
}
});
it("conservatively binds unknown dynamic imports", () => {
const f = fixture(); f.write("selected.ts", "export const load = (name: string) => import(name);");
const before = f.fingerprint(); f.write("unselected.ts", "export const changed = true;");
expect(f.fingerprint()).not.toBe(before);
});
it("executes only the selected domain and rejects metadata drift", async () => {
const f = fixture();
const descriptor: ChatCatalogDescriptor = { scenarioId: "selected", modulePath: "./chat-catalog-domains/selected.js", exportName: "cases",
sourceFiles: ["selected.ts"], oracleVersion: "v1", expectedRunCount: 1, turnTimeoutMs: 100, attemptTimeoutMs: 1000 };
f.write("selected.mjs", `globalThis.__selectedChatLoads = (globalThis.__selectedChatLoads ?? 0) + 1; export const cases = [{...${JSON.stringify(descriptor)}, setup(){}, observe(){}, grade(){}, verifyResult(){}}];`);
f.write("unselected.mjs", 'throw new Error("Unselected domain executed");');
const requested: string[] = [];
const loaded = await loadChatCatalogImplementation(descriptor, async specifier => {
requested.push(specifier);
return import(pathToFileURL(path.join(f.root, "selected.mjs")).href);
});
expect(loaded.scenarioId).toBe("selected"); expect(requested).toHaveLength(1);
expect(requested[0]).toContain("/chat-catalog-domains/selected.js");
await expect(loadChatCatalogImplementation({ ...descriptor, expectedRunCount: 2 }, async () => ({ cases: [loaded] }))).rejects.toThrow("Runtime descriptor drift");
});
it("refuses formerly inactive launch targets and source drift falls back conservatively", () => {
const lane = { suiteId: "chat-features", flow: "agent_chat", environment: "local" };
expect(assertChatCatalogFingerprintLane(lane, {}, [])).toMatchObject({ installedLane: false });
for (const changed of [{ ...lane, environment: "daytona" }, { ...lane, flow: "github_review_journey" }])
expect(() => assertChatCatalogFingerprintLane(changed, {}, [])).toThrow();
expect(() => assertChatCatalogFingerprintLane(lane, { PAPERCLIP_RUNNER_E2E_INSTALLED_CLI: "/candidate" }, [])).toThrow();
expect(() => assertChatCatalogFingerprintLane(lane, {}, ["--target-config"])).toThrow();
const f = fixture(); f.write("selected.ts", "export const load=(name:string)=>import(name);");
const before = fingerprintChatDependencyClosure({ root: f.root, entries: ["selected.ts"], selected: lane,
reviewedDynamicResolutions: { "selected.ts": { sourceSha256: "outdated", resolution: "reviewed-inactive", dependencies: [] } } });
expect(before.unknownDynamicImports).toEqual(["selected.ts"]);
f.write("unselected.ts", "export const changed=true;");
expect(fingerprintChatDependencyClosure({ root: f.root, entries: ["selected.ts"], selected: lane }).digest).not.toBe(before.digest);
});
it("binds selected and spawned code, conservatively closing unresolved CLI and production loaders", () => {
const descriptor = findChatCatalogCase("projects.create-project")!;
const result = chatCatalogCaseFingerprint(descriptor, { suiteId: "chat-features", flow: "agent_chat", environment: "local" });
const files = result.sources.map(source => source.file);
expect(files).toContain("tests/runner-e2e/chat-catalog-domains/internal-projects.ts");
expect(files).toContain("tests/runner-e2e/chat-catalog-domains/internal-support.ts");
expect(files).toContain("packages/adapter-utils/src/workspace-snapshot-thread-worker.ts");
expect(files).toEqual(expect.arrayContaining([...chatCatalogSpawnedSourceEntries, "tests/runner-e2e/chat-catalog-domains/internal-goals.ts"]));
expect(files).toContain("tests/runner-e2e/chat-catalog-domains/identity-auth.ts");
expect(result.unknownDynamicImports).toEqual(expect.arrayContaining(["cli/src/commands/run.ts", "server/src/adapters/plugin-loader.ts"]));
});
it("fails closed when preserved prerequisites or a rejected query lack their independent verifier", async () => {
const base: ChatCatalogDescriptor = { scenarioId: "selected", modulePath: "./chat-catalog-domains/selected.js", exportName: "cases",
sourceFiles: ["selected.ts"], oracleVersion: "v1", expectedRunCount: 1, turnTimeoutMs: 100, attemptTimeoutMs: 1000 };
const callbacks = { setup() {}, observe() {}, grade() {}, verifyResult() {} };
const setup = { ...base, setupPolicy: { funding: "preserve-scenario" as const, invariantCheckIds: ["saved-budget"] } };
await expect(loadChatCatalogImplementation(setup, async () => ({ cases: [{ ...setup, ...callbacks }] }))).rejects.toThrow("independent prerequisite assertions");
const rejected = { ...base, queryOutcome: "rejected" as const };
await expect(loadChatCatalogImplementation(rejected, async () => ({ cases: [{ ...rejected, ...callbacks }] }))).rejects.toThrow("independent rejection evidence");
});
it("replays the retained public name-only identity after raw selected inputs have been removed", () => {
const descriptor = findChatCatalogCase("projects.create-project")!;
const execution = { executionId: "offline-selected", suiteId: "chat-features", flow: "agent_chat", environment: "local" };
const names = { credentialNames: ["OPENAI_API_KEY"], referenceNames: [], externalRequirements: [] };
const before = chatCatalogCaseFingerprint(descriptor, execution, names).digest;
vi.stubEnv("PAPERCLIP_CHAT_PRIVATE_INPUTS", "/removed/owned-secret-file.json");
expect(chatCatalogCaseFingerprint(descriptor, execution, names).digest).toBe(before);
expect(() => chatCatalogCaseFingerprint(descriptor, execution)).toThrow();
});
it("binds extensionless TSX components and directory TSX entries", () => {
const f = fixture(); mkdirSync(path.join(f.root, "nested"));
f.write("selected.ts", 'export { value } from "./Component";');
f.write("Component.tsx", 'export { value } from "./nested";');
f.write("nested/index.tsx", 'export const value = "actual component";');
const before = f.fingerprint();
f.write("nested/index.tsx", 'export const value = "changed component";');
expect(f.fingerprint()).not.toBe(before);
expect(fingerprintChatDependencyClosure({ root: f.root, entries: ["selected.ts"], selected: {} }).sources.map(row => row.file))
.toEqual(expect.arrayContaining(["Component.tsx", "nested/index.tsx"]));
});
it("rejects a missing extensionless component instead of binding an unrelated TSX file", () => {
const f = fixture();
f.write("selected.ts", 'export { value } from "./MissingComponent";');
f.write("OtherComponent.tsx", 'export const value = "unrelated";');
expect(f.fingerprint).toThrow("Unresolved fingerprint dependency: ./MissingComponent");
});
it("keeps the reviewed adapter in the normalized execution an installed OpenCode case is admitted from", async () => {
const { runnerMatrix } = await import("./catalog.js");
const execution = runnerMatrix.find(row => row.id === "chat-features.runner-opencode.local.instance-operations.update-instance");
expect(execution).toBeDefined();
const normalized = normalizedChatCatalogExecution(execution!);
expect(normalized.profile.adapterType).toBe("paperclip_runner");
expect(assertChatCatalogFingerprintLane({ ...normalized, scenarioId: "instance-operations.update-instance", installedSource: "published-global-npm" }, {}, []).installedLane).toBe(true);
});
it("admits installed OpenCode only for its exact reviewed model, case and source-local lane", () => {
const lane = { suiteId: "chat-features", flow: "agent_chat", environment: "local", scenarioId: "instance-operations.update-instance", installedSource: "published-global-npm",
profile: { id: "runner-opencode", provider: "opencode", model: CHAT_OPENCODE_MODEL, adapterType: "paperclip_runner", credential: "OPENROUTER_API_KEY" } };
expect(assertChatCatalogFingerprintLane(lane, {}, []).installedLane).toBe(true);
expect(() => assertChatCatalogFingerprintLane({ ...lane, profile: { ...lane.profile, provider: "codex" } }, {}, [])).toThrow();
expect(() => assertChatCatalogFingerprintLane({ ...lane, scenarioId: "tasks.create-backlog" }, { PAPERCLIP_RUNNER_E2E_INSTALLED_CLI: "/candidate" }, [])).toThrow();
});