mirror of
https://github.com/paperclipai/paperclip.git
synced 2026-11-15 10:21:41 +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>
29 lines
4.9 KiB
TypeScript
29 lines
4.9 KiB
TypeScript
import { describe, expect, it } from "vitest";
|
|
import { gradeNotificationAudience } from "./internal-notification-audience.js";
|
|
import type { InternalObservation } from "./internal-support.js";
|
|
function valid() {
|
|
const t = { id: "task", title: "Launch task", status: "backlog" }; const participation = { id: "human", authorUserId: "alex", body: "Following this draft." };
|
|
const normal = { id: "normal", authorAgentId: "ceo", createdByRunId: "first", body: "The draft is ready for human review." };
|
|
const mention = { id: "mention", authorAgentId: "ceo", createdByRunId: "second", body: "The draft is ready, [Riley](user://riley)." };
|
|
const person = (id: string, eligible: boolean) => ({ principalId: id, visibility: "visible", sockets: ["/api/companies/company/events/ws"], frames: [{ principalId: id, taskId: "task", commentId: "normal", eligible, at: 1000 }], toasts: eligible ? [{ text: "CEO commented on task", hrefs: ["/RUN/issues/task"], at: 1100 }] : [] });
|
|
const first = { scenarioId: "inbox-search.notification-audience", origin: "https://fixture.test", companyId: "company", leadId: "ceo", state: { task: structuredClone(t), alexId: "alex", rileyId: "riley", participation },
|
|
tasks: [t], projects: [], goals: [], comments: [participation, normal], activity: [], documents: [], revisions: [], links: ["/RUN/issues/task"], humanQueries: [], reply: "Posted the draft-ready update.", runs: [{ id: "first", companyId: "company", agentId: "ceo", status: "succeeded" }], alex: person("alex", true), riley: person("riley", false), observedAt: 3100 };
|
|
const result = structuredClone(first) as typeof first & { state: { firstAudience?: typeof first; mentionPhaseStartedAt?: number } };
|
|
result.state.firstAudience = first; result.state.mentionPhaseStartedAt = 4000; result.comments.push(mention); result.runs.push({ id: "second", companyId: "company", agentId: "ceo", status: "succeeded" });
|
|
for (const p of [result.alex, result.riley]) { p.frames.push({ principalId: p.principalId, taskId: "task", commentId: "mention", eligible: true, at: 4500 }); p.toasts.push({ text: "CEO commented on task for Riley", hrefs: ["/RUN/issues/task"], at: 4600 }); }
|
|
result.observedAt = 5000; return result;
|
|
}
|
|
describe("actual per-principal notification calibration", () => {
|
|
it("accepts normal participant-only notification and a separate real mention toast", () => expect(gradeNotificationAudience(valid()).passed).toBe(true));
|
|
it("rejects server eligibility without actual browser delivery", () => { const o = valid(); o.riley.toasts = []; expect(gradeNotificationAudience(o).passed).toBe(false); });
|
|
it("rejects notifying an unrelated reader in the normal phase", () => { const o = valid(); o.state.firstAudience!.riley.frames[0]!.eligible = true; expect(gradeNotificationAudience(o).passed).toBe(false); });
|
|
it("requires the same exact comment event for both principals", () => { const o = valid(); o.riley.frames[1]!.commentId = "another-comment"; expect(gradeNotificationAudience(o).passed).toBe(false); });
|
|
it("rejects an event or toast for a different task", () => { const o = valid(); o.riley.frames[1]!.taskId = "another"; expect(gradeNotificationAudience(o).passed).toBe(false); const wrong = valid(); wrong.riley.toasts[0]!.hrefs = ["/RUN/issues/another"]; expect(gradeNotificationAudience(wrong).passed).toBe(false); });
|
|
it("requires a full bounded negative window and genuine visible sockets", () => { const o = valid(); o.state.firstAudience!.observedAt = 2000; expect(gradeNotificationAudience(o).passed).toBe(false); const hidden = valid(); hidden.riley.visibility = "hidden"; expect(gradeNotificationAudience(hidden).passed).toBe(false); });
|
|
it("rejects plaintext name without an actual user mention", () => { const o = valid(); o.comments[2]!.body = "Riley, the draft is ready."; expect(gradeNotificationAudience(o).passed).toBe(false); });
|
|
it("rejects wrong-phase-only toast or invented browser claims", () => { const o = valid(); o.riley.toasts[0]!.at = 1100; expect(gradeNotificationAudience(o).passed).toBe(false); const claim = valid(); claim.reply += " I confirm the browser notification was delivered."; expect(gradeNotificationAudience(claim).passed).toBe(false); });
|
|
it("accepts a reworded update because the saved event and toasts prove delivery", () => { const o = valid(); o.comments[1]!.body = "Heads up: COM-1 can be looked at now."; o.state.firstAudience!.comments[1]!.body = o.comments[1]!.body; expect(gradeNotificationAudience(o).passed).toBe(true); });
|
|
it("rejects a second normal-phase agent comment", () => { const o = valid(); const extra = { id: "extra", authorAgentId: "ceo", createdByRunId: "first", body: "Another update." }; o.state.firstAudience!.comments.push(extra); expect(gradeNotificationAudience(o).passed).toBe(false); });
|
|
it("accepts a truthful lack of browser-delivery confirmation", () => { const o = valid(); o.reply += " I cannot confirm that the browser notification was delivered."; expect(gradeNotificationAudience(o).passed).toBe(true); });
|
|
});
|