Files
PaperClipAI/tests/runner-e2e/chat-private-inputs.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

104 lines
11 KiB
TypeScript

import { mkdtempSync, writeFileSync, statSync, chmodSync, rmSync } from "node:fs";
import path from "node:path";
import { tmpdir } from "node:os";
import { afterEach, expect, it } from "vitest";
import { stageChatPrivateInputs, readChatPrivateInputs, chatPrivateSetupBlockers, chatLaunchCredentialNames, chatPrivateInputNameIdentity, scrubChatPrivateEnvironment, CHAT_PRIVATE_INPUTS_ENV } from "./chat-private-inputs.js";
import type { MatrixExecution } from "./types.js";
import type { ChatCatalogDescriptor } from "./chat-catalog-registry.js";
import { runtimePrivateRequirements } from "./chat-runtime-private-requirements.js";
const roots: string[] = [];
afterEach(() => { for (const root of roots.splice(0)) rmSync(root, { recursive: true, force: true }); });
function fixture() { const root = mkdtempSync(path.join(tmpdir(), "chat-private-")); roots.push(root); return root; }
const execution = { id: "selected", profile: { credential: "CODEX_AUTH_JSON" }, environment: {} } as MatrixExecution;
it("stages only declared credentials privately and removes them from unrelated process environments", () => {
const root = fixture(), secrets: string[] = [];
const descriptor = { requiredCredentials: ["OPENAI_API_KEY"] } satisfies Pick<ChatCatalogDescriptor, "requiredCredentials" | "externalInput">;
const env = { CODEX_AUTH_JSON: "profile-login", OPENAI_API_KEY: "wizard-value", ANTHROPIC_API_KEY: "other-value", PAPERCLIP_CHAT_EVAL_UNUSED: "unrelated-value" };
const staged = stageChatPrivateInputs({ execution, descriptor, privateDir: root, env, secrets });
expect(statSync(staged.filename).mode & 0o077).toBe(0);
const privateEnv = { [CHAT_PRIVATE_INPUTS_ENV]: staged.filename, PAPERCLIP_RUNNER_E2E_TEMP_ROOT: root };
const inputs = readChatPrivateInputs("selected", secrets, privateEnv);
expect(inputs.credentials.OPENAI_API_KEY).toBe("wizard-value"); expect(inputs.read("ANTHROPIC_API_KEY")).toBeUndefined();
expect(secrets).toContain("wizard-value");
expect(() => readChatPrivateInputs("other", [], privateEnv)).toThrow("another execution");
const child = scrubChatPrivateEnvironment({ ...env, ...privateEnv }, [execution]);
expect(child.CODEX_AUTH_JSON).toBe("profile-login"); expect(child.OPENAI_API_KEY).toBeUndefined(); expect(child.PAPERCLIP_CHAT_EVAL_UNUSED).toBeUndefined();
expect(scrubChatPrivateEnvironment(child, [], true)[CHAT_PRIVATE_INPUTS_ENV]).toBeUndefined();
expect(JSON.stringify(chatPrivateInputNameIdentity("selected", privateEnv))).not.toContain("wizard-value");
chmodSync(staged.filename, 0o644); expect(() => readChatPrivateInputs("selected", [], privateEnv)).toThrow("private0600");
});
it("resolves only the selected provider references and fingerprints names rather than credential values", () => {
const root = fixture(), configFile = path.join(root, "external.json"), secrets: string[] = [];
writeFileSync(configFile, JSON.stringify({ schemaVersion: 1, providers: {
notion: { disposable: true, resourceIds: ["test-page"], credentialEnv: ["PAPERCLIP_CHAT_EVAL_NOTION_AUTH"], prerequisites: { workspaceId: "test-space", tokenEnv: "PAPERCLIP_CHAT_EVAL_NOTION_AUTH", unrelated: "not-selected" } },
slack: { disposable: true, resourceIds: ["unrelated"], credentialEnv: ["PAPERCLIP_CHAT_EVAL_SLACK_AUTH"], prerequisites: {} },
} }), { mode: 0o600 });
const descriptor = { externalInput: { provider: "notion", fields: ["workspaceId", "tokenEnv"] } } satisfies Pick<ChatCatalogDescriptor, "requiredCredentials" | "externalInput">;
const staged = stageChatPrivateInputs({ execution, descriptor, privateDir: root, secrets, env: { PAPERCLIP_CHAT_CATALOG_EXTERNAL_CONFIG: configFile, PAPERCLIP_CHAT_EVAL_NOTION_AUTH: "notion-private", PAPERCLIP_CHAT_EVAL_SLACK_AUTH: "slack-private" } });
const env = { [CHAT_PRIVATE_INPUTS_ENV]: staged.filename, PAPERCLIP_RUNNER_E2E_TEMP_ROOT: root };
const inputs = readChatPrivateInputs("selected", secrets, env);
expect(inputs.read("PAPERCLIP_CHAT_EVAL_NOTION_AUTH")).toBe("notion-private"); expect(inputs.read("PAPERCLIP_CHAT_EVAL_SLACK_AUTH")).toBeUndefined();
expect(JSON.stringify(inputs.configuration("external-provider"))).not.toContain("slack");
expect(JSON.stringify(inputs.configuration("external-provider"))).not.toContain("not-selected");
const identity = JSON.stringify(chatPrivateInputNameIdentity("selected", env));
expect(identity).toContain("PAPERCLIP_CHAT_EVAL_NOTION_AUTH"); expect(identity).not.toContain("notion-private"); expect(identity).not.toContain("test-space");
});
it("stages only an exact declared runtime metadata contract and standard key without dummy provider refs", () => {
const root = fixture(), configFile = path.join(root, "runtime.json");
const scenarioId = "execution-environments.prepare-image"; const spec = runtimePrivateRequirements[scenarioId];
const descriptor = { scenarioId, externalInput: { provider: spec.provider, fields: spec.fields }, requiredCredentials: spec.credentials };
const runtimeExecution = { ...execution, task: { id: scenarioId } } as MatrixExecution;
const prerequisites = { imageDigest: "sha256:synthetic-image-pin", imageRef: "owned-image", region: "test-region", disposableNamespace: "owned-test-namespace", toolsManifest: "private-tools-reference", unrelated: "not-selected" };
writeFileSync(configFile, JSON.stringify({ schemaVersion: 1, providers: { daytona: { disposable: true, resourceIds: ["owned-test-namespace"], credentialEnv: [], prerequisites } } }), { mode: 0o600 });
const staged = stageChatPrivateInputs({ execution: runtimeExecution, descriptor, privateDir: root, secrets: [], env: { PAPERCLIP_CHAT_CATALOG_EXTERNAL_CONFIG: configFile, DAYTONA_API_KEY: "selected-daytona-key", ANTHROPIC_API_KEY: "unrelated-key" } });
const env = { [CHAT_PRIVATE_INPUTS_ENV]: staged.filename, PAPERCLIP_RUNNER_E2E_TEMP_ROOT: root };
const inputs = readChatPrivateInputs("selected", [], env);
expect(inputs.credentials.DAYTONA_API_KEY).toBe("selected-daytona-key"); expect(inputs.credentials.ANTHROPIC_API_KEY).toBeUndefined();
expect(JSON.stringify(inputs.configuration("external-provider"))).not.toContain("not-selected");
const publicIdentity = JSON.stringify(chatPrivateInputNameIdentity("selected", env));
expect(publicIdentity).toContain("imageDigest"); expect(publicIdentity).not.toContain("synthetic-image-pin"); expect(publicIdentity).not.toContain("selected-daytona-key");
expect(() => stageChatPrivateInputs({ execution, descriptor, privateDir: root, secrets: [], env: {} })).toThrow("another scenario");
});
it("rejects undeclared runtime scenarios and fields before resolving or writing private values", () => {
const root = fixture(); const spec = runtimePrivateRequirements["execution-environments.prepare-image"];
for (const descriptor of [
{ scenarioId: "tasks.create-backlog", externalInput: { provider: spec.provider, fields: spec.fields }, requiredCredentials: spec.credentials },
{ scenarioId: "execution-environments.prepare-image", externalInput: { provider: spec.provider, fields: [...spec.fields, "arbitraryCommand"] }, requiredCredentials: spec.credentials },
{ scenarioId: "execution-environments.prepare-image", externalInput: { provider: spec.provider, fields: spec.fields }, requiredCredentials: [] },
]) expect(() => stageChatPrivateInputs({ execution: { ...execution, task: { id: descriptor.scenarioId } } as MatrixExecution, descriptor, privateDir: root, secrets: [], env: {} })).toThrow("exact selected canonical case");
});
it("resolves only the finite SSH private key reference and refuses an alternate ambient name", () => {
const root = fixture(), configFile = path.join(root, "ssh.json"); const scenarioId = "execution-environments.test-computer"; const spec = runtimePrivateRequirements[scenarioId];
const descriptor = { scenarioId, externalInput: { provider: spec.provider, fields: spec.fields }, requiredCredentials: spec.credentials };
const prerequisites = { hostname: "synthetic.test", port: "22", login: "eval", hostKeySha256: "SHA256:synthetic", knownHosts: "synthetic.test ssh-ed25519 synthetic-private-metadata", allowedRemoteRoot: "/owned/disposable", privateKeyEnv: "PAPERCLIP_CHAT_EVAL_SSH_PRIVATE_KEY", qualifiedHarness: "native-codex" };
const config = { schemaVersion: 1, providers: { ssh: { disposable: true, resourceIds: ["owned-ssh"], credentialEnv: [], prerequisites } } };
writeFileSync(configFile, JSON.stringify(config), { mode: 0o600 });
const staged = stageChatPrivateInputs({ execution: { ...execution, task: { id: scenarioId } } as MatrixExecution, descriptor, privateDir: root, secrets: [], env: { PAPERCLIP_CHAT_CATALOG_EXTERNAL_CONFIG: configFile, PAPERCLIP_CHAT_EVAL_SSH_PRIVATE_KEY: "selected-ssh-key" } });
const inputs = readChatPrivateInputs("selected", [], { [CHAT_PRIVATE_INPUTS_ENV]: staged.filename, PAPERCLIP_RUNNER_E2E_TEMP_ROOT: root });
expect(inputs.read("PAPERCLIP_CHAT_EVAL_SSH_PRIVATE_KEY")).toBe("selected-ssh-key"); expect(inputs.read("PAPERCLIP_CHAT_EVAL_OTHER_KEY")).toBeUndefined();
prerequisites.privateKeyEnv = "PAPERCLIP_CHAT_EVAL_OTHER_KEY"; writeFileSync(configFile, JSON.stringify(config), { mode: 0o600 });
rmSync(staged.filename);
const blocked = stageChatPrivateInputs({ execution: { ...execution, task: { id: scenarioId } } as MatrixExecution, descriptor, privateDir: root, secrets: [], env: { PAPERCLIP_CHAT_CATALOG_EXTERNAL_CONFIG: configFile, PAPERCLIP_CHAT_EVAL_OTHER_KEY: "wrong-key" } });
const blockedInputs = readChatPrivateInputs("selected", [], { [CHAT_PRIVATE_INPUTS_ENV]: blocked.filename, PAPERCLIP_RUNNER_E2E_TEMP_ROOT: root });
expect(chatPrivateSetupBlockers(blockedInputs, descriptor)).toContain("unavailable-external-configuration:ssh"); expect(blockedInputs.read("PAPERCLIP_CHAT_EVAL_OTHER_KEY")).toBeUndefined();
});
it("retains missing selected credentials inside a reserved setup receipt while profile credentials stay launch prerequisites", () => {
const root = fixture(); const descriptor = { requiredCredentials: ["OPENAI_API_KEY" as const] };
const selected = { ...execution, requiredCredentials: ["CODEX_AUTH_JSON", "OPENAI_API_KEY"] } as MatrixExecution;
const staged = stageChatPrivateInputs({ execution: selected, descriptor, privateDir: root, secrets: [], env: {} });
const env = { [CHAT_PRIVATE_INPUTS_ENV]: staged.filename, PAPERCLIP_RUNNER_E2E_TEMP_ROOT: root };
const inputs = readChatPrivateInputs("selected", [], env);
expect(chatPrivateSetupBlockers(inputs, descriptor)).toEqual(["missing-credential:OPENAI_API_KEY"]);
expect(chatPrivateInputNameIdentity("selected", env)?.credentialNames).toEqual(["OPENAI_API_KEY"]);
expect(chatLaunchCredentialNames(selected, descriptor)).toEqual(["CODEX_AUTH_JSON"]);
expect(chatLaunchCredentialNames({ ...selected, profile: { ...selected.profile, credential: "OPENAI_API_KEY" } }, descriptor)).toContain("OPENAI_API_KEY");
});
it("stages missing provider fields/configuration as sanitized setup blockers without leaking private filenames", () => {
const root = fixture(); const descriptor = { externalInput: { provider: "notion" as const, fields: ["recentFactJson"] } };
const staged = stageChatPrivateInputs({ execution, descriptor, privateDir: root, secrets: [], env: { PAPERCLIP_CHAT_CATALOG_EXTERNAL_CONFIG: "/private/not-an-existing-fixture" } });
const inputs = readChatPrivateInputs("selected", [], { [CHAT_PRIVATE_INPUTS_ENV]: staged.filename, PAPERCLIP_RUNNER_E2E_TEMP_ROOT: root });
expect(chatPrivateSetupBlockers(inputs, descriptor)).toEqual(["unavailable-external-configuration:notion"]);
expect(JSON.stringify(inputs.setupBlockers)).not.toContain("not-an-existing-fixture");
});