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>
38 lines
5.7 KiB
TypeScript
38 lines
5.7 KiB
TypeScript
import { expect, it, vi } from "vitest";
|
|
import { gradePostStop, sameSourcePostStopChronology, releasePostStopGateAfterCleanup } from "./runtime-post-stop.js";
|
|
function chronology() {
|
|
const old = { id: "old", nativeIssueId: "source", companyId: "company", agentId: "ceo", runtimeMode: "native", status: "cancelled", finishedAt: "2026-10-26T12:00:02Z", resultJson: { nativeCancellation: { schema: "paperclip.native-cancellation.v1", intentId: "intent", intentAuditId: "audit-intent", acknowledgementAuditId: "audit-ack", issueId: "source", companyId: "company", runId: "old", scope: "run", recordedAt: "2026-10-26T12:00:00Z", acknowledgedAt: "2026-10-26T12:00:02Z", dispatchState: "acknowledged", dispatched: true } } };
|
|
return { old, next: { id: "new", nativeIssueId: "source", companyId: "company", agentId: "ceo", runtimeMode: "native", status: "succeeded", startedAt: "2026-10-26T12:00:03Z" }, boundary: { runId: "old", issueId: "source", intentId: "intent", intentRecordedAt: "2026-10-26T12:00:00Z" }, human: { id: "canonical", body: "Now make the website a single page instead.", issueId: "source", companyId: "company", authorAgentId: null, authorUserId: "person", createdAt: "2026-10-26T12:00:01Z" }, query: "Now make the website a single page instead.", companyId: "company", agentId: "ceo", humanId: "person", prompt: "canonical\nNow make the website a single page instead.", wakeCommentIds: ["canonical"], heldReleasedAt: "2026-10-26T12:00:02.500Z" } as unknown as Parameters<typeof sameSourcePostStopChronology>[0];
|
|
}
|
|
function valid() {
|
|
return { companyId: "company", leadId: "ceo", agents: [], links: [], humanQueries: ["initial", "canonical"], reply: "", tasks: [], activity: [], runs: [{ id: "old", agentId: "ceo", status: "cancelled" }, { id: "new", agentId: "ceo", status: "succeeded" }], state: { stopStatus: 200, messageAcceptedDuringHeldCleanup: true, noSuccessorWhileHeld: true, gateReleasedAfterCleanup: true, chronologyBound: true, providerExited: true, initialHumanRetained: true, singleSuccessor: true, sameCurrentAssignment: true, oldEvidenceRetained: true, actualSimplePage: true, artifactRunBound: true, canonicalReply: "The single-page website is built and saved here." } };
|
|
}
|
|
it("binds the real unchanged canonical human instruction between same-source Stop intent and the end of held cleanup", () => expect(sameSourcePostStopChronology(chronology())).toBe(true));
|
|
it("rejects wrong-source, early, late, duplicate, resumed or unacknowledged deliveries", () => {
|
|
for (const change of [(v: ReturnType<typeof chronology>) => { v.human.issueId = "other"; }, (v: ReturnType<typeof chronology>) => { v.old.nativeIssueId = "other"; }, (v: ReturnType<typeof chronology>) => { v.next.agentId = "other"; }, (v: ReturnType<typeof chronology>) => { v.human.createdAt = "2026-10-26T12:00:03Z" as never; }, (v: ReturnType<typeof chronology>) => { v.next.startedAt = "2026-10-26T12:00:02.400Z" as never; }, (v: ReturnType<typeof chronology>) => { v.next.startedAt = "2026-10-26T12:00:01Z" as never; }, (v: ReturnType<typeof chronology>) => { v.wakeCommentIds.push("initial"); }, (v: ReturnType<typeof chronology>) => { v.prompt = "a different instruction"; }, (v: ReturnType<typeof chronology>) => { v.old.status = "succeeded"; }, (v: ReturnType<typeof chronology>) => { (v.old.resultJson!.nativeCancellation as Record<string, unknown>).dispatchState = "pending"; }]) { const v = chronology(); change(v); expect(sameSourcePostStopChronology(v)).toBe(false); }
|
|
});
|
|
it("accepts real post-cleanup delivery and a natural substantive result paraphrase", () => {
|
|
expect(gradePostStop(valid()).passed).toBe(true); const o = valid(); o.state.canonicalReply = "Acorn Notes is ready, with its signup action and all the content together. The artifact is saved in this conversation."; expect(gradePostStop(o).passed).toBe(true);
|
|
});
|
|
it("rejects polite completion without actual same-source cleanup, delivery or rendered output", () => {
|
|
for (const key of ["messageAcceptedDuringHeldCleanup", "noSuccessorWhileHeld", "gateReleasedAfterCleanup", "chronologyBound", "providerExited", "initialHumanRetained", "singleSuccessor", "sameCurrentAssignment", "oldEvidenceRetained", "actualSimplePage", "artifactRunBound"] as const) { const o = valid(); o.state[key] = false; expect(gradePostStop(o).passed).toBe(false); }
|
|
const o = valid(); o.state.canonicalReply = ""; expect(gradePostStop(o).passed).toBe(false); expect(gradePostStop(undefined).passed).toBe(false);
|
|
});
|
|
|
|
it("releases the neutral command gate after real cleanup without waiting for successor success", async () => {
|
|
const release = vi.fn(async () => {}), old = chronology().old;
|
|
await expect(releasePostStopGateAfterCleanup(old, false, release)).rejects.toThrow("cleanup");
|
|
await expect(releasePostStopGateAfterCleanup({ ...old, status: "running" }, true, release)).rejects.toThrow("cleanup");
|
|
const pending = chronology().old; (pending.resultJson!.nativeCancellation as Record<string, unknown>).dispatchState = "pending";
|
|
await expect(releasePostStopGateAfterCleanup(pending, true, release)).rejects.toThrow("cleanup"); expect(release).not.toHaveBeenCalled();
|
|
// No successor outcome is an input: it may still be running and retrying the
|
|
// legitimate initial command. Releasing only its neutral result avoids a cycle.
|
|
await releasePostStopGateAfterCleanup(old, true, release); expect(release).toHaveBeenCalledOnce();
|
|
});
|
|
|
|
it("accepts a message saved after the bounded Stop ACK while the old provider is still held in cleanup", () => {
|
|
// The product ACKs Stop after a bounded cleanup grace; the provider may still be closing then.
|
|
const v = chronology(); v.human.createdAt = "2026-10-26T12:00:02.200Z" as never; expect(sameSourcePostStopChronology(v)).toBe(true);
|
|
v.human.createdAt = "2026-10-26T12:00:02.600Z" as never; expect(sameSourcePostStopChronology(v)).toBe(false);
|
|
});
|