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>
94 lines
8.0 KiB
TypeScript
94 lines
8.0 KiB
TypeScript
import { expect } from "@playwright/test";
|
|
import type { Request } from "@playwright/test";
|
|
import { checks } from "./internal-support.js";
|
|
import { privilegedDefinition, privilegedObservation, settlePrivilegedChat } from "./privileged-support.js";
|
|
|
|
function fixtureBoundary(reply: string) {
|
|
const positive = reply.replace(/\b(?:not|isn't|does not|cannot|can't)\b[^.!?\n]*/gi, "");
|
|
return /synthetic|fixture|design reference|presentation|storybook|mock/i.test(reply)
|
|
&& !/not\s+(?:a\s+)?(?:synthetic|fixture|mock)/i.test(reply)
|
|
&& !/real\s+(?:authorization|continuation)\s+proof|(?:verified|proved|confirmed).{0,35}(?:real|live|production).{0,25}(?:authorization|denial|continuation|workflow)/i.test(positive);
|
|
}
|
|
export function gradeComponentReference(value: unknown) {
|
|
const c = checks(), o = privilegedObservation(value);
|
|
c.add("component-observation", !!o, "Actual browser state transitions and the run-attributed canonical reply are required."); if (!o) return c.result();
|
|
c.add("existing-question-reference", /design-guide|DesignGuide|TaskPendingInputExample|TaskChatComposer|storybook/i.test(o.reply)
|
|
&& /question|decision|pending input/i.test(o.reply) && /dismiss|reopen|resolv|answer|pending/i.test(o.reply),
|
|
"The reply identifies an existing design or Storybook question example and relevant states.");
|
|
c.add("question-states-opened", typeof o.state.referencePath === "string" && /^\/[^/]+\/design-guide$/.test(o.state.referencePath)
|
|
&& o.state.referenceCompanyId === o.companyId && o.state.questionVisible === true
|
|
&& o.state.questionDismissed === true && o.state.questionReopened === true && o.state.questionResolved === true,
|
|
"The actual supported design guide renders the question and its dismiss, reopen and resolution states.");
|
|
c.add("reference-only", fixtureBoundary(o.reply) && o.state.fixtureMutations === 0 && o.state.beforeRuns === o.state.afterRuns,
|
|
"The fixture does not perform API mutations or admit native work; rendering is not authorization or continuation proof.");
|
|
return c.result();
|
|
}
|
|
export function gradeInteractionFixture(value: unknown) {
|
|
const c = checks(), o = privilegedObservation(value);
|
|
c.add("denial-presentation-observation", !!o, "A real browser must display the existing synthetic denial surface."); if (!o) return c.result();
|
|
c.add("existing-denial-lab-guidance", /responsible-user-denial|ResponsibleUserDenialUxLab|denial.{0,30}(?:lab|fixture)|(?:lab|fixture).{0,30}denial/i.test(o.reply)
|
|
&& fixtureBoundary(o.reply), "The response identifies the existing denial lab and explains its synthetic limitation.");
|
|
c.add("denial-and-action-rendered", o.state.referencePath === "/ux-lab/responsible-user-denial"
|
|
&& o.state.denialVisible === true && o.state.denialActionVisible === true && o.state.unavailableStateVisible === true,
|
|
"The actual denial notice displays both the blocked action and the supported human recovery guidance.");
|
|
c.add("no-real-tool-call", o.state.fixtureMutations === 0 && o.state.beforeRuns === o.state.afterRuns,
|
|
"Opening this presentation fixture issues no mutation and admits no real model/tool work.");
|
|
return c.result();
|
|
}
|
|
|
|
/** No fixture data is inserted into the conversation or substituted for model output. */
|
|
export const privilegedDeveloperLabCases = [
|
|
privilegedDefinition("developer-labs.component-reference", "privileged-developer-labs", {
|
|
async setup() { return {}; }, grade: gradeComponentReference,
|
|
async act(ctx, state) {
|
|
await settlePrivilegedChat(ctx);
|
|
const source = ctx.input.page.url(), page = ctx.input.page;
|
|
let writes = 0; const observe = (request: Request) => { if (new URL(request.url()).pathname.startsWith("/api/") && !["GET", "HEAD", "OPTIONS"].includes(request.method())) writes++; };
|
|
state.beforeRuns = (await ctx.allRuns()).length;
|
|
page.on("request", observe);
|
|
try {
|
|
await page.goto(new URL(`/${ctx.input.fixtures.company.issuePrefix}/design-guide`, ctx.input.api.baseURL).href, { waitUntil: "domcontentloaded" });
|
|
const section = page.locator("section").filter({ has: page.getByRole("heading", { name: "Pending task input above composer", exact: true }) });
|
|
await expect(section.getByText("Should the agent use the existing draft?", { exact: true })).toBeVisible(); state.questionVisible = true;
|
|
await ctx.input.capture("question-component-open", "Synthetic question card open", "question-component-open.png");
|
|
await section.getByRole("button", { name: "Dismiss Question", exact: true }).click();
|
|
await expect(section.getByTestId("task-chat-composer-takeover")).toHaveCount(0); state.questionDismissed = true;
|
|
await section.getByTestId("task-chat-pending-input-indicator").click();
|
|
await expect(section.getByText("Should the agent use the existing draft?", { exact: true })).toBeVisible(); state.questionReopened = true;
|
|
await section.getByRole("button", { name: "Use draft", exact: true }).click();
|
|
await expect(section.getByTestId("task-chat-composer-takeover")).toHaveCount(0);
|
|
await expect(section.getByTestId("task-chat-pending-input-indicator")).toHaveCount(0); state.questionResolved = true;
|
|
state.referencePath = new URL(page.url()).pathname;
|
|
state.referenceCompanyId = ctx.input.fixtures.company.id;
|
|
await ctx.input.capture("question-component-reference", "Synthetic question card reference states", "question-component-reference.png");
|
|
} finally { page.off("request", observe); state.fixtureMutations = writes; await page.goto(source, { waitUntil: "domcontentloaded" }); }
|
|
state.afterRuns = (await ctx.allRuns()).length;
|
|
await ctx.input.evidence("chat-catalog-question-reference.json", { ...state, scope: "presentation-only; no authorization or continuation proof" });
|
|
},
|
|
async verifyResult(ctx) { await expect(ctx.input.page.getByTestId("task-chat-composer-input")).toBeVisible(); },
|
|
}),
|
|
privilegedDefinition("developer-labs.interaction-fixture", "privileged-developer-labs", {
|
|
async setup() { return {}; }, grade: gradeInteractionFixture,
|
|
async act(ctx, state) {
|
|
await settlePrivilegedChat(ctx);
|
|
const page = ctx.input.page, source = page.url(); let writes = 0;
|
|
const observe = (request: Request) => { if (new URL(request.url()).pathname.startsWith("/api/") && !["GET", "HEAD", "OPTIONS"].includes(request.method())) writes++; };
|
|
state.beforeRuns = (await ctx.allRuns()).length; page.on("request", observe);
|
|
try {
|
|
await page.goto(new URL("/ux-lab/responsible-user-denial", ctx.input.api.baseURL).href, { waitUntil: "domcontentloaded" });
|
|
const denial = page.locator('[data-testid="responsible-user-denial-notice"][data-denial-code="RESPONSIBLE_USER_UNAUTHORIZED"]').first();
|
|
await expect(denial.getByText("Responsible user not authorized", { exact: true })).toBeVisible(); state.denialVisible = true;
|
|
await expect(denial).toContainText("have someone who is authorized take this action instead"); state.denialActionVisible = true;
|
|
await expect(page.locator('[data-denial-code="RESPONSIBLE_USER_UNAVAILABLE"]').first()).toContainText("Mark the work blocked"); state.unavailableStateVisible = true;
|
|
state.referencePath = new URL(page.url()).pathname;
|
|
await ctx.input.capture("synthetic-denial-reference", "Synthetic denial and human action guidance", "synthetic-denial-reference.png");
|
|
} finally { page.off("request", observe); state.fixtureMutations = writes; await page.goto(source, { waitUntil: "domcontentloaded" }); }
|
|
state.afterRuns = (await ctx.allRuns()).length;
|
|
await ctx.input.evidence("chat-catalog-denial-reference.json", { ...state, scope: "presentation-only; no real provider denial" });
|
|
},
|
|
async verifyResult(ctx) { await expect(ctx.input.page.getByTestId("task-chat-composer-input")).toBeVisible(); },
|
|
}),
|
|
];
|
|
for (const definition of privilegedDeveloperLabCases) definition.sourceFiles = [...definition.sourceFiles,
|
|
"../../ui/src/App.tsx", "../../ui/src/pages/DesignGuide.tsx", "../../ui/src/pages/ResponsibleUserDenialUxLab.tsx"];
|