mirror of
https://github.com/paperclipai/paperclip.git
synced 2026-11-14 00:26:28 +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>
56 lines
5.7 KiB
TypeScript
56 lines
5.7 KiB
TypeScript
import { createAuthenticatedPerson } from "./identity-auth.js";
|
|
import type { RunnerApi } from "../api.js";
|
|
import type { Browser } from "@playwright/test";
|
|
import { it, expect, vi, afterEach } from "vitest";
|
|
import { gradeConferenceRoomOff, probeBoardChatAsPerson, type ConferenceRoomOffObservation } from "./runtime-chat-disabled.js";
|
|
function valid(): ConferenceRoomOffObservation { return { companyId: "company", companyPrefix: "RUN", agentId: "morgan-id", personId: "actual-human",
|
|
canonicalIntent: "Open the conference room so I can talk with Morgan.", conferenceRoomEnabled: false, oldLinkLandedPath: "/RUN/dashboard",
|
|
conferenceComposerVisible: false, conferenceNavOffered: false, chatAlternativePath: "/RUN/chats/morgan", chatAlternativeAgentRefs: ["morgan-id", "morgan"],
|
|
chatComposerVisible: true, boardChatStatus: 403, boardChatErrorCode: "FEATURE_DISABLED", boardChatProbePersonId: "actual-human",
|
|
agentSettingsStatus: 403, gateAfterProbe: false, beforeTaskIds: ["seed"], afterTasks: [{ id: "seed", conversationAgentId: null }], runIds: [] }; }
|
|
const fails = (change: Partial<ConferenceRoomOffObservation>) => expect(gradeConferenceRoomOff({ ...valid(), ...change }).passed).toBe(false);
|
|
it("accepts the off Conference Room landing on the dashboard with Chat as the working alternative", () => expect(gradeConferenceRoomOff(valid()).passed).toBe(true));
|
|
it("accepts Morgan's own ordinary chat record opened from Chat", () => expect(gradeConferenceRoomOff({ ...valid(), chatAlternativePath: "/RUN/chats/morgan-id",
|
|
afterTasks: [{ id: "seed", conversationAgentId: null }, { id: "morgan-chat", conversationAgentId: "morgan-id" }] }).passed).toBe(true));
|
|
it("rejects missing context or the experiment actually on", () => { expect(gradeConferenceRoomOff(undefined).passed).toBe(false); fails({ conferenceRoomEnabled: true }); });
|
|
it("rejects a broken route, an unusable conference composer or a lingering nav entry", () => {
|
|
for (const change of [{ oldLinkLandedPath: "/RUN/board-chat" }, { oldLinkLandedPath: "/OTHER/dashboard" }, { conferenceComposerVisible: true }, { conferenceNavOffered: true }]) fails(change);
|
|
});
|
|
it("rejects a missing or wrong chat alternative", () => {
|
|
for (const change of [{ chatComposerVisible: false }, { chatAlternativePath: "/RUN/chats/someone-else" }, { chatAlternativePath: "/OTHER/chats/morgan" }]) fails(change);
|
|
});
|
|
it("rejects a board discussion that is not refused by the Conference Room gate", () => {
|
|
for (const change of [{ boardChatStatus: 200 }, { boardChatErrorCode: "DEPLOYMENT_MODE_UNSUPPORTED" }, { boardChatErrorCode: "trusted-origin-required" }, { boardChatProbePersonId: "someone-else" }]) fails(change);
|
|
});
|
|
it("rejects an ordinary agent enabling the instance setting", () => fails({ agentSettingsStatus: 200, gateAfterProbe: true }));
|
|
it("rejects a disguised task, a removed record or started native work", () => {
|
|
fails({ afterTasks: [{ id: "seed", conversationAgentId: null }, { id: "board-operations", conversationAgentId: null }] });
|
|
fails({ afterTasks: [{ id: "seed", conversationAgentId: null }, { id: "other-chat", conversationAgentId: "other-agent" }] });
|
|
fails({ afterTasks: [] }); fails({ runIds: ["unrequested"] });
|
|
});
|
|
|
|
afterEach(() => vi.unstubAllGlobals());
|
|
it("probes with the real person's session cookie and trusted origin, and classifies only the closed refusal code", async () => {
|
|
const origin = "http://127.0.0.1:3100", captured: Array<{ url: string; method: string; headers: Headers; body: string }> = [];
|
|
vi.stubGlobal("fetch", vi.fn(async (input: URL | string, options: RequestInit = {}) => {
|
|
const url = new URL(String(input)), headers = new Headers(options.headers);
|
|
if (url.pathname === "/api/auth/sign-up/email") return new Response(JSON.stringify({ user: { id: "actual-human", name: "Alex" } }), { status: 200, headers: { "set-cookie": "fixture-session=unit-private-session; Path=/; HttpOnly" } });
|
|
if (url.pathname === "/api/auth/sign-out") return new Response("{}", { status: 200 });
|
|
captured.push({ url: url.href, method: options.method ?? "GET", headers, body: String(options.body ?? "") });
|
|
const correct = headers.get("origin") === origin && headers.get("cookie") === "fixture-session=unit-private-session" && !headers.has("authorization");
|
|
return new Response(JSON.stringify(correct ? { error: "Conference Room Chat is not enabled", code: "FEATURE_DISABLED" } : { error: "Board mutation requires trusted browser origin" }), { status: 403 });
|
|
}));
|
|
const context = { setStorageState: vi.fn(async () => {}), close: vi.fn(async () => {}) };
|
|
const person = await createAuthenticatedPerson({ api: { baseURL: origin, get: async () => ({ deploymentMode: "authenticated" }) } as unknown as RunnerApi,
|
|
browser: { newContext: async () => context } as unknown as Browser, secrets: [], name: "Alex" });
|
|
try {
|
|
expect(await probeBoardChatAsPerson(person, "owned-company", "Open the conference room so I can talk with Morgan."))
|
|
.toEqual({ boardChatStatus: 403, boardChatErrorCode: "FEATURE_DISABLED", boardChatProbePersonId: "actual-human" });
|
|
expect(captured).toHaveLength(1); expect(captured[0]!.url).toBe(`${origin}/api/board/chat/stream`); expect(captured[0]!.method).toBe("POST");
|
|
expect(JSON.parse(captured[0]!.body)).toEqual({ companyId: "owned-company", message: "Open the conference room so I can talk with Morgan." });
|
|
const missingOrigin = await probeBoardChatAsPerson({ id: person.id, request: async path => fetch(new URL(path, origin), { method: "POST", headers: { Cookie: "fixture-session=unit-private-session" } }) }, "owned-company", "x");
|
|
expect(missingOrigin).toMatchObject({ boardChatStatus: 403, boardChatErrorCode: "trusted-origin-required" });
|
|
expect(gradeConferenceRoomOff({ ...valid(), ...missingOrigin }).passed).toBe(false);
|
|
} finally { await person.dispose(); }
|
|
});
|