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>
17 lines
5.3 KiB
TypeScript
17 lines
5.3 KiB
TypeScript
import { it, expect } from "vitest";
|
|
import { genuineGrokRoundTrip, truthfulGrokCapabilities, gradeExistingGrok, supportedGrokHumanHandoff } from "./runtime-grok.js";
|
|
function connection() { return { enabled: true, resourceUrl: "https://owned.invalid/mcp/runner", agentLifecycleState: "ready", recoveryNeeded: false, binding: { id: "binding", providerKind: "grok_bot", connected: true, wakeVerified: true, routine: { bindingId: "binding", botName: "Owned existing Bot", configured: true, verified: true, recoveryNeeded: false, verifiedAt: "2026-10-27T01:02:00Z", lastContactAt: "2026-10-27T01:02:00Z", lastExchange: { id: "challenge", kind: "verify", status: "acknowledged", sentAt: "2026-10-27T01:01:00Z", acknowledgedAt: "2026-10-27T01:02:00Z" } } } } as unknown as NonNullable<Parameters<typeof genuineGrokRoundTrip>[0]>; }
|
|
function valid() { return { companyId: "company", leadId: "ceo", agents: [{ id: "bot" }], links: [], tasks: [], activity: [], humanQueries: ["Connect my existing Grok Bot as an agent here.", "I used the supported Grok Bot setup and sent the verification challenge. Check the saved connection and tell me whether it is ready and what capabilities are unavailable."], runs: [{ id: "initial", agentId: "ceo", status: "succeeded" }, { id: "result", agentId: "ceo", status: "succeeded" }], reply: "", state: { beforeAgentIds: [] as string[], enabledPublicInstance: true, noExistingBotAuthorization: true, authorizationUiVisible: true, pairingSecretStayedPrivate: true, initialReply: "Use the supported Grok Bot setup; authorize Paperclip in your browser and connect your existing Bot.", operatorAvailable: true, humanHandoffCompleted: true, publicPersonGrantBound: true, actualHumanPairingObserved: true, sameChosenCompanyAndBot: true, realRoundTrip: true, routineConfiguredSecurely: true, worker: { id: "bot", companyId: "company", adapterType: "paperclip_runner", adapterConfig: { provider: "grok_bot" }, lifecycleState: "ready" }, confirmationBound: true, resultReply: "The authenticated round trip is verified and the Bot is ready. Usage and spending are not reported. Workspace files remain external, and Stop cannot prove that external execution terminated." } }; }
|
|
it("accepts actual authenticated same-Bot verification and truthful natural capability limitations", () => { expect(genuineGrokRoundTrip(connection(), "binding", "Owned existing Bot")).toBe(true); expect(gradeExistingGrok(valid()).passed).toBe(true); expect(truthfulGrokCapabilities("Connected and verified. Paperclip cannot report usage or cost, cannot inspect the Bot's workspace, and cannot guarantee remote termination.")).toBe(true); });
|
|
it("rejects provider acceptance without actual authenticated acknowledgement, wrong Bot/binding and ready flags alone", () => {
|
|
const good = connection(); expect(genuineGrokRoundTrip(good, "foreign", "Owned existing Bot")).toBe(false); expect(genuineGrokRoundTrip(good, "binding", "Different Bot")).toBe(false);
|
|
for (const mutate of [(v: typeof good) => { v.binding!.routine!.lastExchange!.status = "accepted"; }, (v: typeof good) => { v.binding!.wakeVerified = false; }, (v: typeof good) => { v.binding!.connected = false; }, (v: typeof good) => { v.binding!.routine!.lastExchange!.acknowledgedAt = null; }, (v: typeof good) => { v.binding!.routine!.recoveryNeeded = true; }, (v: typeof good) => { v.resourceUrl = "http://localhost/mcp/runner"; }]) { const v = connection(); mutate(v); expect(genuineGrokRoundTrip(v, "binding", "Owned existing Bot")).toBe(false); }
|
|
});
|
|
it("rejects simulated readiness, changed choice, wrong phase, local parity and zero-cost guesses", () => {
|
|
for (const mutate of [(o: ReturnType<typeof valid>) => { o.state.realRoundTrip = false; }, (o: ReturnType<typeof valid>) => { o.state.actualHumanPairingObserved = false; }, (o: ReturnType<typeof valid>) => { o.state.sameChosenCompanyAndBot = false; }, (o: ReturnType<typeof valid>) => { o.state.confirmationBound = false; }, (o: ReturnType<typeof valid>) => { o.state.initialReply = "I've connected your Grok Bot; it is now ready."; }, (o: ReturnType<typeof valid>) => { o.state.resultReply += " It has the same capabilities as a local coding agent and guarantees remote termination."; }]) { const o = valid(); mutate(o); expect(gradeExistingGrok(o).passed).toBe(false); }
|
|
expect(truthfulGrokCapabilities("Usage costs $0. Workspace is external and execution termination is not guaranteed.")).toBe(false); expect(gradeExistingGrok(undefined).passed).toBe(false);
|
|
});
|
|
|
|
it("requires an actual supported human handoff and never rescues a silent or already-completed setup claim", () => { expect(supportedGrokHumanHandoff("Open Agents, choose New Agent and Invite an external agent, then select Grok Bot and complete authorization in your browser.")).toBe(true); expect(supportedGrokHumanHandoff("You can connect your existing Bot from its profile; use your browser to authorize it and verify the routine.")).toBe(true); expect(supportedGrokHumanHandoff("I've connected the Grok Bot and it is ready.")).toBe(false); expect(supportedGrokHumanHandoff("Grok is an external system.")).toBe(false); });
|
|
it("rejects operator acknowledgement alone without actual public-person grant and authenticated Bot proof", () => { for (const key of ["operatorAvailable", "humanHandoffCompleted", "publicPersonGrantBound", "realRoundTrip"]) { const o = valid(); (o.state as Record<string, unknown>)[key] = false; expect(gradeExistingGrok(o).passed).toBe(false); } });
|