mirror of
https://github.com/paperclipai/paperclip.git
synced 2026-11-15 18:56:05 +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>
41 lines
5.2 KiB
TypeScript
41 lines
5.2 KiB
TypeScript
import { describe, expect, it } from "vitest";
|
|
import { gradeIntegratorCaseClose } from "./integrator-case-close.js";
|
|
function valid() {
|
|
const original = { id: "case", workIssueId: "primary", status: "draft", completedAt: null };
|
|
const review = { id: "review", issueId: "primary", latestRevisionId: "reviewed", body: "Accepted: correction permits activation once and retains the Pay name." };
|
|
const base = { scenarioId: "cases.close-case", origin: "https://fixture.test", companyId: "company", leadId: "ceo", state: { case: original, task: { id: "primary", status: "done" }, worker: { id: "morgan" }, review, reviewRunId: "review-native", firstReplyTime: "2026-10-26T10:00:00Z" },
|
|
tasks: [{ id: "primary", status: "done" }], projects: [], goals: [], documents: [], revisions: [], comments: [], links: [], activity: [], humanQueries: [],
|
|
caseRecord: { ...original, status: "done", completedAt: "2026-10-26T10:02:00Z" }, handoffs: [{ destinationKind: "done", state: "completed", issueId: "primary", actorType: "agent", actorId: "morgan", completedAt: "2026-10-26T10:02:00Z", sourceRunId: "terminal-native" }],
|
|
runs: [{ id: "terminal-native", agentId: "morgan", status: "succeeded", runtimeMode: "native", contextSnapshot: { issueId: "primary" } }],
|
|
finalDocument: review, finalRevisions: [{ id: "reviewed", body: review.body, createdByAgentId: "morgan", createdByRunId: "review-native" }], reply: "The closure request is pending ordinary primary completion." };
|
|
const result = { ...structuredClone(base), reply: `${base.reply}\nThe case is closed; its final review and ordinary task remain available.` };
|
|
return { ...base, reply: `${result.reply}\nThe other case cannot close: its required review is unresolved.`, negativeCase: { id: "negative", workIssueId: "secondary", status: "draft", completedAt: null },
|
|
negativeTask: { id: "secondary", status: "blocked" }, pendingReview: { id: "required", status: "backlog" }, blockerEvidence: { blockers: [{ id: "required", isUnresolved: true }] },
|
|
state: { ...base.state, initialDirection: structuredClone(base), acceptedResult: result } };
|
|
}
|
|
describe("explicit case terminal completion with normal review gates", () => {
|
|
it("accepts actual primary terminal receipt, later factual result and distinct unresolved review refusal", () => {
|
|
expect(gradeIntegratorCaseClose(valid()).passed).toBe(true); expect(gradeIntegratorCaseClose(undefined).passed).toBe(false);
|
|
const natural = valid(); natural.reply = `${natural.state.acceptedResult.reply}\nThe separate case is not yet closed; we must complete its required review first.`;
|
|
expect(gradeIntegratorCaseClose(natural).passed).toBe(true);
|
|
});
|
|
it("rejects task-done-only, standing-chat terminal authority and fixture preclosure", () => {
|
|
const taskOnly = valid(); taskOnly.handoffs = []; expect(gradeIntegratorCaseClose(taskOnly).passed).toBe(false);
|
|
const chat = valid(); chat.runs[0]!.contextSnapshot.issueId = "standing-chat"; expect(gradeIntegratorCaseClose(chat).passed).toBe(false);
|
|
const before = valid(); before.state.case.status = "done"; expect(gradeIntegratorCaseClose(before).passed).toBe(false);
|
|
const incomplete = valid(); incomplete.handoffs[0]!.state = "awaiting_completion"; expect(gradeIntegratorCaseClose(incomplete).passed).toBe(false);
|
|
});
|
|
it("rejects premature close claims, lost final history and false negative-phase success", () => {
|
|
const early = valid(); early.state.initialDirection.reply = "The case is closed."; expect(gradeIntegratorCaseClose(early).passed).toBe(false);
|
|
const lost = valid(); lost.finalRevisions = []; expect(gradeIntegratorCaseClose(lost).passed).toBe(false);
|
|
const gate = valid(); gate.negativeCase.status = "done"; expect(gradeIntegratorCaseClose(gate).passed).toBe(false);
|
|
const fake = valid(); fake.reply = `${fake.state.acceptedResult.reply}\nThe review is complete and the case is closed.`; expect(gradeIntegratorCaseClose(fake).passed).toBe(false);
|
|
const contradictory = valid(); contradictory.reply = `${contradictory.state.acceptedResult.reply}\nThe case is closed; its review is still pending.`; expect(gradeIntegratorCaseClose(contradictory).passed).toBe(false);
|
|
const concealed = valid(); concealed.state.initialDirection.reply = "The case is closed; its review is awaiting final approval."; expect(gradeIntegratorCaseClose(concealed).passed).toBe(false);
|
|
const negated = valid(); negated.reply = `${negated.state.acceptedResult.reply}\nThe case is not closed: its review is pending.`; expect(gradeIntegratorCaseClose(negated).passed).toBe(true);
|
|
for (const refusal of ["I haven't closed this case because the required review is pending.", "I have not closed the case; the required review is unresolved."]) { const truthful = valid(); truthful.reply = `${truthful.state.acceptedResult.reply}\n${refusal}`; expect(gradeIntegratorCaseClose(truthful).passed).toBe(true); }
|
|
const past = valid(); past.state.acceptedResult.reply = `${past.state.initialDirection.reply}\nI closed the case.`; past.reply = `${past.state.acceptedResult.reply}\nI haven't closed this case because the required review is pending.`; expect(gradeIntegratorCaseClose(past).passed).toBe(true);
|
|
const noGate = valid(); noGate.blockerEvidence.blockers = []; expect(gradeIntegratorCaseClose(noGate).passed).toBe(false);
|
|
});
|
|
});
|