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>
40 lines
4.9 KiB
TypeScript
40 lines
4.9 KiB
TypeScript
import { it, expect } from "vitest";
|
|
import { gradeGovernedHire } from "./runtime-hiring.js";
|
|
function replyLink(commentId: string, href: string) {
|
|
return { threadTestId: "task-chat-thread" as const, href, source: "reply" as const, conversationId: "source", anchorCommentId: commentId, commentId };
|
|
}
|
|
function valid() {
|
|
const proposal = (id: string, name: string) => ({ id, companyId: "company", type: "hire_agent", status: "pending", requestedByAgentId: "ceo", decidedByUserId: null as string | null,
|
|
payload: { agentId: name.toLowerCase(), name, role: "qa", adapterType: "runner", adapterConfig: { model: "admitted" } } });
|
|
const firstProposal = proposal("first", "Morgan"), secondProposal = proposal("second", "Jules");
|
|
const acceptedBefore = { id: "jules", companyId: "company", name: "Jules", role: "qa", adapterType: "runner", adapterConfig: { model: "admitted" } };
|
|
return { scenarioId: "agents.hire-needs-approval", origin: "http://fixture.invalid", companyId: "company", leadId: "ceo",
|
|
state: { firstProposal, secondProposal, approvalsAfter: [{ ...firstProposal, status: "rejected", decidedByUserId: "human" }, { ...secondProposal, status: "approved", decidedByUserId: "human" }],
|
|
secondRequestAttributed: true, rejectedBeforeChangedRequest: true, noRepeatedProposalBeforeChangedRequest: true, acceptedBefore,
|
|
noWorkerBeforeFirstDecision: true, noWorkerBeforeSecondDecision: true, requestRunIds: ["canonical", "changed"],
|
|
firstReply: "Morgan's QA proposal is pending board approval.", secondReply: "Jules's QA proposal is awaiting approval.",
|
|
firstReplyCommentIds: ["reply-1"], secondReplyCommentIds: ["reply-2"] },
|
|
renderedLinkEvidence: [replyLink("reply-1", "/RUN/approvals/first"), replyLink("reply-1", "/RUN/agents/morgan"), replyLink("reply-2", "/RUN/approvals/second")],
|
|
agents: [{ ...acceptedBefore, status: "idle", lifecycleState: "ready" }], tasks: [], links: [], activity: [], humanQueries: [], reply: "",
|
|
runs: [{ id: "canonical", agentId: "ceo", status: "succeeded", contextSnapshot: {} as Record<string, unknown> }, { id: "changed", agentId: "ceo", status: "succeeded", contextSnapshot: {} as Record<string, unknown> }] };
|
|
}
|
|
it("accepts distinct real rejection and acceptance branches after an explicit changed request", () => expect(gradeGovernedHire(valid()).passed).toBe(true));
|
|
it("does not grade reply wording: a reworded reply with the same approval links passes", () => { const o = valid(); o.state.firstReply = "Morgan is set up as a QA hire; open the approval below."; o.state.secondReply = "Here is Jules. Decide on the card."; expect(gradeGovernedHire(o).passed).toBe(true); });
|
|
it("rejects missing governance evidence", () => expect(gradeGovernedHire(undefined).passed).toBe(false));
|
|
it("rejects a working employee before the decision", () => { const o = valid(); o.state.noWorkerBeforeFirstDecision = false; expect(gradeGovernedHire(o).passed).toBe(false); });
|
|
it("rejects repeating the rejected proposal without changed human direction", () => { const o = valid(); o.state.noRepeatedProposalBeforeChangedRequest = false; expect(gradeGovernedHire(o).passed).toBe(false); });
|
|
it("rejects accepting a replacement identity instead of the proposed hire", () => { const o = valid(); o.agents[0]!.id = "replacement"; expect(gradeGovernedHire(o).passed).toBe(false); });
|
|
it("rejects a request reply that does not link its own pending approval", () => {
|
|
const o = valid(); o.renderedLinkEvidence = o.renderedLinkEvidence.filter(link => link.href !== "/RUN/approvals/second"); expect(gradeGovernedHire(o).passed).toBe(false);
|
|
const swapped = valid(); swapped.state.secondReplyCommentIds = ["reply-1"]; expect(gradeGovernedHire(swapped).passed).toBe(false);
|
|
const foreign = valid(); foreign.renderedLinkEvidence = [replyLink("reply-1", "/RUN/approvals/first"), replyLink("reply-2", "https://elsewhere.invalid/RUN/approvals/second")]; expect(gradeGovernedHire(foreign).passed).toBe(false);
|
|
const api = valid(); api.renderedLinkEvidence = [replyLink("reply-1", "/RUN/approvals/first"), replyLink("reply-2", "/api/approvals/second")]; expect(gradeGovernedHire(api).passed).toBe(false);
|
|
});
|
|
it("rejects a hidden nudge classified as a decision notification", () => { const o = valid(); o.runs.push({ id: "extra", agentId: "ceo", status: "succeeded", contextSnapshot: { source: "approval.approved", approvalId: "foreign" } }); expect(gradeGovernedHire(o).passed).toBe(false); });
|
|
it("accepts a separately counted exact approval notification", () => { const o = valid(); o.runs.push({ id: "extra", agentId: "ceo", status: "succeeded", contextSnapshot: { source: "approval.approved", approvalId: "second" } }); expect(gradeGovernedHire(o).passed).toBe(true); });
|
|
|
|
it("accepts an absolute same-origin approval link", () => {
|
|
const o = valid(); o.renderedLinkEvidence = [replyLink("reply-1", "http://fixture.invalid/RUN/approvals/first"), replyLink("reply-2", "/RUN/approvals/second")];
|
|
expect(gradeGovernedHire(o).passed).toBe(true);
|
|
});
|