mirror of
https://github.com/paperclipai/paperclip.git
synced 2026-11-14 08:53:19 +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
5.3 KiB
TypeScript
56 lines
5.3 KiB
TypeScript
import { afterEach, expect, it, vi } from "vitest";
|
|
import type { ChatCatalogContext } from "../chat-catalog-contract.js";
|
|
vi.mock("@playwright/test", () => ({ expect: (target: any) => ({
|
|
toBeVisible: async () => { if (!await target.visible()) throw Error("Actual Chat is still loading or its conversation is absent"); },
|
|
toBeEditable: async () => { if (!target.editable()) throw Error("Chat composer is not ready"); },
|
|
toHaveURL: async (matches: (url: URL) => boolean) => { if (!matches(new URL(target.url()))) throw Error("Wrong selected Chat route"); },
|
|
}) }));
|
|
import { runtimePrimaryCases } from "./runtime-primary.js";
|
|
afterEach(() => { vi.restoreAllMocks(); });
|
|
function fixture(options: { loadingForever?: boolean; extraRun?: boolean; extraChat?: boolean; wrongPrimary?: boolean; wrongPerson?: boolean;
|
|
wrongChat?: boolean; noReply?: boolean; wrongRequest?: boolean; missingRenderedReply?: boolean; readOnly?: boolean; expired?: boolean; lateReady?: boolean } = {}) {
|
|
const deadline = Date.now() + (options.expired ? -1 : 30_000); let current = "https://fixture.test/RUN/chats/ceo", ready = false;
|
|
const calls: string[] = [];
|
|
const source = { id: "source", companyId: "company", conversationAgentId: "ceo", conversationUserId: "person" };
|
|
const comments = [{ id: "human", authorUserId: "person", body: "Make Morgan my go-to agent." },
|
|
...(options.noReply ? [] : [{ id: "reply", authorAgentId: "ceo", createdByRunId: "native", body: "Use the human primary control." }])];
|
|
const primary = { companyId: "company", userId: options.wrongPerson ? "foreign" : "person", primaryAgentId: options.wrongPrimary ? "ceo" : "morgan" };
|
|
const locator = (kind: string): any => ({ first() { return this; }, locator: () => locator("editor"), getByRole: () => locator("heading"),
|
|
visible: async () => {
|
|
calls.push(kind);
|
|
if (kind === "composer") {
|
|
if (options.loadingForever) return false; ready = true;
|
|
if (options.lateReady) vi.spyOn(Date, "now").mockReturnValue(deadline + 1);
|
|
}
|
|
return kind === "heading" || ready && !(kind.includes("comment-reply") && options.missingRenderedReply);
|
|
}, editable: () => ready && !options.readOnly });
|
|
const page = { url: () => current, goto: async (url: string) => { current = new URL(url, "https://fixture.test").href; ready = false; },
|
|
reload: async () => {}, getByRole: () => locator("heading"), getByTestId: () => locator("composer"), locator };
|
|
const api = { baseURL: "https://fixture.test", get: async (path: string) => {
|
|
calls.push(path); if (path.endsWith("/primary-agent/me")) return primary;
|
|
if (path.endsWith("/chats/ceo")) return { ...source, ...(options.wrongChat ? { id: "different-source" } : {}) };
|
|
if (path.includes("/comments?")) return comments.map(row => options.wrongRequest && row.id === "human" ? { ...row, body: "different query" } : row);
|
|
if (path.endsWith("/chats")) return [source, ...(options.extraChat && ready ? [{ ...source, id: "extra" }] : [])];
|
|
throw Error("Unexpected API read");
|
|
} };
|
|
const state = { worker: { id: "morgan", name: "Morgan", companyId: "company" }, primaryBefore: { companyId: "company", userId: "person", primaryAgentId: "ceo" },
|
|
sourceBeforeNavigation: "source", nativeIdsBeforeNavigation: ["native"], chatCountBeforeNavigation: 1 };
|
|
const evidence = vi.fn(async (_name: string, _value: unknown) => {});
|
|
const ctx = { observationDeadlineAt: deadline, scenario: { id: "agents.choose-primary", query: "Make Morgan my go-to agent." },
|
|
input: { api, page, evidence, fixtures: { company: { id: "company", issuePrefix: "RUN" }, agent: { id: "ceo" } } },
|
|
issue: () => source, refreshIssue: async () => {}, allRuns: async () => [{ id: "native", agentId: "ceo", status: "succeeded" }, ...(options.extraRun && ready ? [{ id: "unexpected", agentId: "ceo", status: "running" }] : [])],
|
|
} as unknown as ChatCatalogContext;
|
|
const observation = { companyId: "company", leadId: "ceo", state, reply: "Use the human primary control.", agents: [{ id: "ceo", name: "CEO" }, state.worker], tasks: [], links: [], activity: [], runs: [], humanQueries: [] };
|
|
return { ctx, state, observation, evidence, calls, ready: () => ready };
|
|
}
|
|
it("actual final verifier waits for meaningful selected Chat then rechecks preference and no extra work before capture", async () => {
|
|
const f = fixture(); await runtimePrimaryCases[0]!.verifyResult(f.ctx, f.state, f.observation);
|
|
expect(f.ready()).toBe(true); expect(f.evidence).toHaveBeenCalledWith("chat-primary-ready-conversation.json", expect.objectContaining({ sourceId: "source", agentId: "ceo", primaryAgentId: "morgan", noExtraWork: true }));
|
|
expect(f.calls).toContain('[id="comment-human"]'); expect(f.calls).toContain('[id="comment-reply"]');
|
|
});
|
|
it.each([{ loadingForever: true }, { extraRun: true }, { extraChat: true }, { wrongPrimary: true }, { wrongPerson: true }, { wrongChat: true },
|
|
{ noReply: true }, { wrongRequest: true }, { missingRenderedReply: true }, { readOnly: true }, { expired: true }, { lateReady: true }])("actual final verifier rejects incomplete or changed settled state %j", async options => {
|
|
const f = fixture(options); await expect(runtimePrimaryCases[0]!.verifyResult(f.ctx, f.state, f.observation)).rejects.toThrow();
|
|
expect(f.evidence.mock.calls.some(row => row[0] === "chat-primary-ready-conversation.json")).toBe(false);
|
|
});
|