mirror of
https://github.com/paperclipai/paperclip.git
synced 2026-11-13 07:12:00 +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>
95 lines
7.6 KiB
TypeScript
95 lines
7.6 KiB
TypeScript
// @vitest-environment jsdom
|
|
import { afterEach, beforeEach, expect, it, vi } from "vitest";
|
|
import { linkedQuestionBinding, readLinkedQuestionDom } from "./chat-linked-question.js";
|
|
import { valid, dom, type Input } from "./chat-linked-question.test-fixture.js";
|
|
|
|
beforeEach(() => { document.body.innerHTML = dom(); vi.spyOn(Element.prototype, "getClientRects").mockReturnValue([{ width: 20, height: 20 }] as unknown as DOMRectList); });
|
|
afterEach(() => vi.restoreAllMocks());
|
|
const collect = (input = valid()) => { const bound = linkedQuestionBinding(input); return bound && readLinkedQuestionDom(bound); };
|
|
it("collects an actual production question takeover's two visible saved-task links with no ordinary reply", () => {
|
|
expect(document.querySelectorAll('[id^="comment-"] a')).toHaveLength(0);
|
|
const evidence = collect();
|
|
expect(evidence?.links).toEqual(["/RUN/issues/RUN-1", "/RUN/issues/experiment-task"]);
|
|
expect(evidence?.text).toBe("Which Landing page task should be High priority?");
|
|
expect(evidence).toMatchObject({ interactionId: "question", questionId: "which-task", anchorCommentId: "human", sourceRunId: "run", leadId: "ceo" });
|
|
});
|
|
it("keeps supported unprefixed issue hrefs with exact saved identities and current company page binding", () => {
|
|
const input = valid();
|
|
input.interactions[0]!.payload.questionSet!.description = "Choose [Launch](/issues/launch-task) or [Experiments](/issues/experiment-task).";
|
|
const anchors = document.querySelectorAll("a");
|
|
anchors[0]!.setAttribute("href", "/issues/launch-task"); anchors[1]!.setAttribute("href", "/issues/experiment-task");
|
|
expect(collect(input)?.links).toEqual(["/issues/launch-task", "/issues/experiment-task"]);
|
|
for (const pageUrl of ["https://fixture.test/OTHER/chats/ceo", "https://foreign.test/RUN/chats/ceo", "https://fixture.test/chats/ceo"]) {
|
|
input.pageUrl = pageUrl; expect(collect(input)).toBeUndefined();
|
|
}
|
|
});
|
|
it.each([
|
|
"Choose [Launch][launch] or [Experiments][experiments].\n\n[launch]: /RUN/issues/RUN-1\n[experiments]: /RUN/issues/experiment-task",
|
|
"Choose [Launch](/RUN/issues/RUN-1) or [Experiments](/RUN/issues/experiment-task). Context: [Launch project](/RUN/projects/launch).",
|
|
])("reads actual rendered task hrefs without a second Markdown spelling or context policy: %s", description => {
|
|
const input = valid(); input.interactions[0]!.payload.questionSet!.description = description;
|
|
document.querySelector('[aria-label="Question context"]')!.insertAdjacentHTML("beforeend", '<a href="/RUN/projects/launch">Launch project</a><a href="https://docs.example.test/context">Context</a>');
|
|
expect(collect(input)?.links).toEqual(["/RUN/issues/RUN-1", "/RUN/issues/experiment-task"]);
|
|
});
|
|
it.each([
|
|
["wrong company", (v: Input) => { v.interactions[0]!.companyId = "foreign"; }],
|
|
["wrong conversation", (v: Input) => { v.interactions[0]!.issueId = "other-source"; }],
|
|
["wrong CEO", (v: Input) => { v.interactions[0]!.createdByAgentId = "another-agent"; }],
|
|
["human-created card", (v: Input) => { v.interactions[0]!.createdByUserId = "person"; }],
|
|
["wrong run", (v: Input) => { v.interactions[0]!.sourceRunId = "old-run"; }],
|
|
["stale card", (v: Input) => { v.interactions[0]!.createdAt = "2026-10-27T12:00:00Z"; }],
|
|
["future card", (v: Input) => { v.interactions[0]!.createdAt = "2026-10-29T12:00:00Z"; }],
|
|
["resolved card", (v: Input) => { v.interactions[0]!.status = "accepted"; }],
|
|
["dismissed card", (v: Input) => { v.interactions[0]!.dismissedAt = "2026-10-28T12:00:40Z"; }],
|
|
["wrong addressed person", (v: Input) => { v.interactions[0]!.addresseeUserId = "other"; }],
|
|
["agent addressed", (v: Input) => { v.interactions[0]!.addresseeAgentId = "ceo"; }],
|
|
["invalid resolver", (v: Input) => { v.interactions[0]!.effectiveResolverPolicy = "invalid" as never; }],
|
|
["wrong origin question", (v: Input) => { v.interactions[0]!.originCommentIds = ["old-human"]; }],
|
|
["foreign origin comment", (v: Input) => { v.interactions[0]!.originCommentIds!.push("foreign-comment"); }],
|
|
["wrong source comment", (v: Input) => { v.interactions[0]!.sourceCommentId = "other-human"; }],
|
|
["missing create action", (v: Input) => { v.observation.activity = []; }],
|
|
["wrong create actor", (v: Input) => { v.observation.activity[0]!.actorId = "other"; }],
|
|
["wrong create company", (v: Input) => { v.observation.activity[0]!.companyId = "foreign"; }],
|
|
["wrong action run", (v: Input) => { Object.assign(v.observation.activity[0]!, { runId: "old-run" }); }],
|
|
["wrong action resource kind", (v: Input) => { Object.assign(v.observation.activity[0]!, { entityType: "project" }); }],
|
|
["wrong action question", (v: Input) => { v.observation.activity[0]!.details!.interactionId = "other"; }],
|
|
["wrong action kind", (v: Input) => { v.observation.activity[0]!.details!.interactionKind = "request_confirmation"; }],
|
|
["legacy run", (v: Input) => { v.observation.runs[0]!.runtimeMode = "legacy"; }],
|
|
["unrelated run issue", (v: Input) => { v.observation.runs[0]!.contextSnapshot = { issueId: "other" }; }],
|
|
["foreign saved task", (v: Input) => { v.tasks[0]!.companyId = "foreign"; }],
|
|
["wrong actual task", (v: Input) => { v.observation.tasks[0]!.id = "other"; }],
|
|
["duplicate saved task", (v: Input) => { v.tasks[1] = v.tasks[0]!; }],
|
|
["old human request", (v: Input) => { v.comments[0]!.body = "another query"; }],
|
|
["nonhuman original request", (v: Input) => { v.comments[0]!.authorAgentId = "ceo"; }],
|
|
["missing linked context", (v: Input) => { v.interactions[0]!.payload.questionSet!.description = ""; }],
|
|
["another native action", (v: Input) => { v.interactions.push(structuredClone(v.interactions[0]!)); }],
|
|
] as const)("rejects %s metadata rather than any question-card presence", (_, change) => {
|
|
const input = valid(); change(input); expect(collect(input)).toBeUndefined();
|
|
});
|
|
it.each([
|
|
["foreign company", "/OTHER/issues/RUN-1"], ["foreign origin", "https://foreign.test/RUN/issues/RUN-1"], ["wrong task", "/RUN/issues/wrong"],
|
|
["task search", "/RUN/issues?search=RUN-1"], ["fragment", "/RUN/issues/RUN-1#unrelated"],
|
|
] as const)("rejects %s saved question link", (_, href) => {
|
|
const input = valid(); input.interactions[0]!.payload.questionSet!.description = input.interactions[0]!.payload.questionSet!.description!.replace("/RUN/issues/RUN-1", href);
|
|
document.querySelector("a")!.setAttribute("href", href);
|
|
expect(collect(input)).toBeUndefined();
|
|
});
|
|
const domNegatives: Array<[string, () => void]> = [
|
|
["missing exact interaction", () => document.querySelector("button")!.id = "interaction-other"],
|
|
["wrong visible question", () => document.querySelector("p")!.textContent = "Another question?"],
|
|
["wrong question binding", () => document.querySelector("p")!.id = "question-other-prompt"],
|
|
["hidden takeover", () => document.querySelector("section")!.hidden = true],
|
|
["hidden prompt", () => document.querySelector("p")!.setAttribute("aria-hidden", "true")],
|
|
["hidden linked task", () => document.querySelector("a")!.setAttribute("style", "display:none")],
|
|
["no actual context links", () => document.querySelectorAll("a").forEach(node => node.remove())],
|
|
["wrong DOM link", () => document.querySelector("a")!.setAttribute("href", "/RUN/issues/wrong")],
|
|
["sidebar link only", () => document.body.append(document.querySelector("a")!)],
|
|
["unrelated conversation", () => document.querySelector("main")!.querySelector("#comment-human")!.remove()],
|
|
["ambiguous conversation", () => document.body.insertAdjacentHTML("beforeend", dom())],
|
|
["unrelated takeover", () => document.querySelector("section")!.setAttribute("data-testid", "another-takeover")],
|
|
];
|
|
it.each(domNegatives)("rejects %s DOM evidence", (_, change) => {
|
|
change();
|
|
expect(collect()).toBeUndefined();
|
|
});
|