mirror of
https://github.com/paperclipai/paperclip.git
synced 2026-11-13 15:55:46 +01:00
## 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>
90 lines
6.6 KiB
TypeScript
90 lines
6.6 KiB
TypeScript
import { afterEach, beforeEach, describe, expect, it, vi, type MockInstance } from "vitest";
|
|
import type { ChatCatalogContext } from "../chat-catalog-contract.js";
|
|
const install = vi.hoisted(() => vi.fn(async (context: { restoreOwner(): void }) => context.restoreOwner()));
|
|
vi.mock("./identity-auth.js", () => ({ installIdentitySession: install }));
|
|
vi.mock("../api.js", () => ({ RunnerApi: class {
|
|
baseURL = "http://127.0.0.1:3100";
|
|
setBrowserSession() {}
|
|
async getRunObservation() { return []; }
|
|
async get() { return {}; }
|
|
} }));
|
|
import { installCommentaryReader } from "./internal-commentary-reader.js";
|
|
|
|
function fixture() {
|
|
let apiUser = "owner"; let browserUser = "owner";
|
|
let source: { id: string; companyId: string; conversationUserId: string; conversationAgentId: string } | undefined;
|
|
const state = { browserOverride: undefined as string | undefined, sourceOverride: undefined as string | undefined };
|
|
let response: Response;
|
|
let cancel: MockInstance<ReadableStream["cancel"]>;
|
|
const transport = vi.fn(async () => {
|
|
response = Response.json({ user: { id: state.browserOverride ?? browserUser } });
|
|
cancel = vi.spyOn(response.body!, "cancel");
|
|
return response;
|
|
});
|
|
vi.stubGlobal("fetch", transport);
|
|
const context = { cookies: vi.fn(async () => [{ name: "session", value: browserUser === "owner" ? "owned-test-cookie" : "reader-test-cookie" }]),
|
|
request: { get: vi.fn(async () => response!) }, restoreOwner() { browserUser = "owner"; } };
|
|
const api = { baseURL: "http://127.0.0.1:3100", get: vi.fn(async () => ({ user: { id: apiUser } })),
|
|
setBrowserSession: vi.fn(() => { apiUser = "owner"; }),
|
|
post: vi.fn(async () => source = { id: "source", companyId: "11111111-1111-4111-8111-111111111111", conversationUserId: state.sourceOverride ?? apiUser, conversationAgentId: "lead" }) };
|
|
const evidence = vi.fn(async () => undefined);
|
|
const page = { context: () => context, reload: vi.fn(async () => undefined) };
|
|
const ctx = { scenario: { id: "activity.agent-feedback" }, input: { api, page, evidence, fixtures: { company: { id: "11111111-1111-4111-8111-111111111111" }, agent: { id: "lead" } } },
|
|
refreshIssue: vi.fn(async () => undefined), issue: () => source } as unknown as ChatCatalogContext;
|
|
const person = { id: "casey", installSessionOn: vi.fn(async () => { browserUser = "casey"; }),
|
|
bindApiSession: vi.fn(() => { apiUser = "casey"; }) } as unknown as Parameters<typeof installCommentaryReader>[1];
|
|
return { ctx, person, api, context, state, evidence, transport, get response() { return response!; }, get cancel() { return cancel!; }, current: () => ({ apiUser, browserUser }) };
|
|
}
|
|
beforeEach(() => { install.mockClear(); });
|
|
afterEach(() => { vi.unstubAllGlobals(); });
|
|
describe("real commentary reader identity lifetime", () => {
|
|
it("aligns real API/browser/conversation identities and restores the original owner only once", async () => {
|
|
const f = fixture(); const selected = await installCommentaryReader(f.ctx, f.person, "owner");
|
|
expect(f.current()).toEqual({ apiUser: "casey", browserUser: "casey" });
|
|
expect(selected.receipt).toEqual({ ownerId: "owner", readerId: "casey", apiUserId: "casey", browserUserId: "casey", sourceIssueId: "source", sourceConversationUserId: "casey" });
|
|
expect(JSON.stringify(f.evidence.mock.calls)).not.toMatch(/owned-test-cookie|session=/);
|
|
expect(fetch).toHaveBeenCalledWith(new URL("/api/auth/get-session", f.api.baseURL), expect.objectContaining({ headers: { Cookie: "session=reader-test-cookie", Origin: f.api.baseURL } }));
|
|
await selected.restore(); await selected.restore();
|
|
expect(f.current()).toEqual({ apiUser: "owner", browserUser: "owner" });
|
|
expect(install).toHaveBeenCalledOnce(); expect(f.cancel).not.toHaveBeenCalled();
|
|
expect(f.response.bodyUsed).toBe(true); expect(f.response.body?.locked).toBe(true);
|
|
});
|
|
it("restores genuine owner sessions after a browser/API identity mismatch", async () => {
|
|
const f = fixture(); f.state.browserOverride = "different-reader";
|
|
await expect(installCommentaryReader(f.ctx, f.person, "owner")).rejects.toThrow("same real reader");
|
|
expect(f.current()).toEqual({ apiUser: "owner", browserUser: "owner" });
|
|
expect(f.evidence).not.toHaveBeenCalled();
|
|
});
|
|
it("rejects a conversation owned by a different person, retaining no false identity receipt", async () => {
|
|
const f = fixture(); f.state.sourceOverride = "owner";
|
|
await expect(installCommentaryReader(f.ctx, f.person, "owner")).rejects.toThrow("actual reader");
|
|
expect(f.current()).toEqual({ apiUser: "owner", browserUser: "owner" });
|
|
expect(f.evidence).not.toHaveBeenCalled();
|
|
});
|
|
it("disposes an unread denied browser response without leaking its body or creating a reader", async () => {
|
|
const f = fixture(); const denied = new Response("private denial body must not escape", { status: 403 });
|
|
const cancel = vi.spyOn(denied.body!, "cancel"); f.transport.mockImplementation(async () => denied);
|
|
const error = await installCommentaryReader(f.ctx, f.person, "owner").catch(error => error);
|
|
expect(error.message).toBe("Commentary browser session could not be independently read");
|
|
expect(error.message).not.toContain("private denial"); expect(cancel).toHaveBeenCalledOnce();
|
|
expect(f.current()).toEqual({ apiUser: "owner", browserUser: "owner" }); expect(f.evidence).not.toHaveBeenCalled();
|
|
});
|
|
it("preserves a malformed successful JSON failure instead of masking it with locked-stream cancellation", async () => {
|
|
const f = fixture(); const malformed = new Response("invalid JSON");
|
|
const cancel = vi.spyOn(malformed.body!, "cancel"); f.transport.mockImplementation(async () => malformed);
|
|
await expect(installCommentaryReader(f.ctx, f.person, "owner")).rejects.toBeInstanceOf(SyntaxError);
|
|
expect(cancel).not.toHaveBeenCalled(); expect(f.current()).toEqual({ apiUser: "owner", browserUser: "owner" });
|
|
expect(f.evidence).not.toHaveBeenCalled();
|
|
});
|
|
it("rejects substituting the existing private owner for a distinct excluded reader", async () => {
|
|
const f = fixture(); f.person.id = "owner";
|
|
await expect(installCommentaryReader(f.ctx, f.person, "owner")).rejects.toThrow("distinct real owner and reader");
|
|
expect(f.person.installSessionOn).not.toHaveBeenCalled(); expect(install).not.toHaveBeenCalled();
|
|
});
|
|
it("requires a genuine restorable owner browser session before switching", async () => {
|
|
const f = fixture(); f.context.cookies.mockResolvedValue([]);
|
|
await expect(installCommentaryReader(f.ctx, f.person, "owner")).rejects.toThrow("genuine browser session");
|
|
expect(f.person.installSessionOn).not.toHaveBeenCalled();
|
|
});
|
|
});
|