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>
57 lines
4.7 KiB
TypeScript
57 lines
4.7 KiB
TypeScript
import { describe, expect, it, vi } from "vitest";
|
|
import { chatRunEvidenceApi, createOwnedChatRunEvidenceObserver, readChatRuns } from "./chat-run-evidence-observer.js";
|
|
import type { ChatFlowInput } from "./chat-flow.js";
|
|
const companyId = "11111111-1111-4111-8111-111111111111", runId = "22222222-2222-4222-8222-222222222222", foreign = "33333333-3333-4333-8333-333333333333";
|
|
function fixture() {
|
|
const row = { id: runId, companyId, agentId: "worker", status: "succeeded" };
|
|
const owner = { baseURL: "http://127.0.0.1:3100", get: vi.fn(async () => ({ content: "actual owned log" })),
|
|
getRunObservation: vi.fn(async (path: string): Promise<unknown> => path.includes("/companies/") ? [row] : path.includes("/events?") ? [] : row) };
|
|
const observer = createOwnedChatRunEvidenceObserver(owner as Parameters<typeof createOwnedChatRunEvidenceObserver>[0], companyId);
|
|
const actor = { baseURL: owner.baseURL, get: vi.fn(async () => { throw Object.assign(new Error("Private source denied"), { status: 403 }); }), getRunObservation: vi.fn(async (path: string): Promise<unknown> => path.includes("/companies/") ? [row] : row) };
|
|
const input = { api: actor, fixtures: { company: { id: companyId } }, execution: { suite: { id: "chat-features" }, task: { id: "activity.agent-feedback" } }, ownedRunEvidenceObserver: observer } as unknown as ChatFlowInput;
|
|
return { owner, observer, row, actor, input };
|
|
}
|
|
describe("exact commentary case independent run evidence", () => {
|
|
it("uses the privately frozen owner only for GET evidence, preserving the actual product actor", async () => {
|
|
const f = fixture(); expect(await readChatRuns(f.input)).toEqual([f.row]);
|
|
await f.observer.getRunObservation(`/api/heartbeat-runs/${runId}/events?afterSeq=0&limit=1000`);
|
|
await f.observer.get(`/api/heartbeat-runs/${runId}/log?limitBytes=1048576`);
|
|
expect(f.owner.get).toHaveBeenCalledExactlyOnceWith(`/api/heartbeat-runs/${runId}/log?limitBytes=1048576`);
|
|
expect(f.actor.getRunObservation).not.toHaveBeenCalled(); expect(f.input.api).toBe(f.actor);
|
|
await expect(f.input.api.get(`/api/issues/${foreign}`)).rejects.toMatchObject({ status: 403 });
|
|
expect(Object.keys(f.observer).sort()).toEqual(["companyId", "origin", "listRuns", "get", "getRunObservation"].sort());
|
|
expect(Object.isFrozen(f.observer)).toBe(true);
|
|
});
|
|
it("keeps ordinary/default run collection on the unchanged product API", async () => {
|
|
const f = fixture(); delete f.input.ownedRunEvidenceObserver;
|
|
expect(chatRunEvidenceApi(f.input)).toBe(f.actor); expect(await readChatRuns(f.input)).toEqual([f.row]);
|
|
expect(f.owner.getRunObservation).not.toHaveBeenCalled();
|
|
});
|
|
it.each(["scenario", "company", "origin"]) ("rejects an observer borrowed into another %s", kind => {
|
|
const f = fixture();
|
|
if (kind === "scenario") f.input.execution.task.id = "activity.work-timeline";
|
|
if (kind === "company") f.input.fixtures.company.id = foreign;
|
|
if (kind === "origin") Object.assign(f.actor, { baseURL: "https://other.test" });
|
|
expect(() => chatRunEvidenceApi(f.input)).toThrow("exact approved scenario/company/origin");
|
|
});
|
|
it("rejects a run not independently listed in the owned company", async () => {
|
|
const f = fixture(); await expect(f.observer.get(`/api/heartbeat-runs/${runId}`)).rejects.toThrow("independently listed");
|
|
await f.observer.listRuns(); await expect(f.observer.get(`/api/heartbeat-runs/${foreign}`)).rejects.toThrow("independently listed");
|
|
});
|
|
it.each(["https://foreign.test/api/heartbeat-runs/", "/api/issues/", "/api/heartbeat-runs/"]) ("rejects unauthorized origin/path %s", async prefix => {
|
|
const f = fixture(); await f.observer.listRuns();
|
|
const path = prefix + runId + (prefix === "/api/heartbeat-runs/" ? "/cancel" : "");
|
|
await expect(f.observer.get(path)).rejects.toThrow();
|
|
});
|
|
it("rejects foreign/duplicate company listings and mismatched detail identity", async () => {
|
|
const f = fixture(); f.row.companyId = foreign; await expect(f.observer.listRuns()).rejects.toThrow("foreign");
|
|
const duplicate = fixture(); duplicate.owner.getRunObservation.mockResolvedValue([duplicate.row, duplicate.row]);
|
|
await expect(duplicate.observer.listRuns()).rejects.toThrow("duplicate");
|
|
const wrong = fixture(); await wrong.observer.listRuns(); wrong.row.companyId = foreign;
|
|
await expect(wrong.observer.get(`/api/heartbeat-runs/${runId}`)).rejects.toThrow("owned list identity");
|
|
});
|
|
it.each(["/events?afterSeq=0&limit=5&mutation=1", "/log?limitBytes=99999999", "?includeSecrets=1", "/events?afterSeq=0&limit=5#fragment"])("rejects unauthorized evidence query %s", async suffix => {
|
|
const f = fixture(); await f.observer.listRuns(); await expect(f.observer.get(`/api/heartbeat-runs/${runId}${suffix}`)).rejects.toThrow();
|
|
});
|
|
});
|