mirror of
https://github.com/paperclipai/paperclip.git
synced 2026-11-14 00:26:28 +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>
61 lines
5.0 KiB
TypeScript
61 lines
5.0 KiB
TypeScript
import { it, expect } from "vitest";
|
|
import { gradeDependencyCycle } from "./runtime-task-coordination.js";
|
|
function valid() {
|
|
const website = { id: "website", title: "Website", companyId: "company", status: "backlog", blockedBy: [] };
|
|
const design = { id: "design", title: "Design", companyId: "company", status: "backlog", blockedBy: [{ id: "website" }] };
|
|
return { scenarioId: "task-coordination.reject-dependency-cycle", origin: "http://fixture.invalid", companyId: "company", leadId: "ceo",
|
|
state: { task: website, design }, dependencyTasks: [structuredClone(website), structuredClone(design)], agents: [], tasks: [website, design],
|
|
links: ["/ACORN/issues/website", "/ACORN/issues/design"], reply: "Design already depends on Website. That would be circular. Choose which should go first.",
|
|
activity: [], runs: [{ id: "run" }], humanQueries: [], conversationInteractions: [] as Array<Record<string, unknown>> };
|
|
}
|
|
/** The pending card retained from run 42678089404, which ended the turn with an empty reply. */
|
|
function card() {
|
|
return { id: "card", kind: "ask_user_questions", status: "pending", companyId: "company", createdByAgentId: "ceo", sourceRunId: "run",
|
|
resolverPolicy: "human_only", requestedResolverPolicy: "human_only", effectiveResolverPolicy: "human_only",
|
|
title: "Choose dependency order",
|
|
summary: "Design (COM-2) currently waits for Website (COM-1). Paperclip won't let Website also wait for Design because that would be a dependency cycle, so which order should I keep?",
|
|
payload: { version: 1, questions: [{ id: "order", prompt: "Which dependency order should I set?", selectionMode: "single",
|
|
options: [{ id: "website_waits_design", label: "Website waits for Design" }, { id: "keep_design_waits_website", label: "Keep current order" }] }] } };
|
|
}
|
|
function withCard(edit: (row: ReturnType<typeof card>) => void = () => {}) {
|
|
const o = valid(); o.reply = ""; const row = card(); edit(row); o.conversationInteractions = [row]; return o;
|
|
}
|
|
it("accepts an independently saved acyclic rejection", () => expect(gradeDependencyCycle(valid()).passed).toBe(true));
|
|
it("accepts natural actionable cycle explanation", () => { const o = valid(); o.reply = "Design already waits for Website. Adding the reverse edge would make a dependency cycle; tell me which task should go first."; expect(gradeDependencyCycle(o).passed).toBe(true); });
|
|
it("rejects missing dependency detail", () => expect(gradeDependencyCycle(undefined).passed).toBe(false));
|
|
it("rejects a plausible explanation with the cycle actually saved", () => {
|
|
const o = valid(); o.dependencyTasks[0]!.blockedBy = [{ id: "design" }]; expect(gradeDependencyCycle(o).passed).toBe(false);
|
|
});
|
|
it("rejects silently deleting the original prerequisite", () => {
|
|
const o = valid(); o.dependencyTasks[1]!.blockedBy = []; expect(gradeDependencyCycle(o).passed).toBe(false);
|
|
});
|
|
it("rejects a conflict with no actionable order decision", () => {
|
|
const o = valid(); o.reply = "Website and Design form a cycle."; expect(gradeDependencyCycle(o).passed).toBe(false);
|
|
});
|
|
|
|
it("accepts a saved acyclic rejection naming both tasks without mandatory chat hrefs", () => {
|
|
const o = valid(); o.links = []; expect(gradeDependencyCycle(o).passed).toBe(true);
|
|
});
|
|
|
|
it("accepts this run's pending order card when the turn ends on it", () => expect(gradeDependencyCycle(withCard()).passed).toBe(true));
|
|
it("accepts a reworded order card naming both tasks", () => expect(gradeDependencyCycle(withCard(row => {
|
|
row.title = "Dependency order"; row.summary = "Website can't wait for Design while Design waits for Website.";
|
|
row.payload.questions[0]!.prompt = "What should happen?";
|
|
row.payload.questions[0]!.options = [{ id: "a", label: "Design first" }, { id: "b", label: "Swap so Website goes after Design" }];
|
|
})).passed).toBe(true));
|
|
it("rejects an empty reply with no card", () => { const o = valid(); o.reply = ""; expect(gradeDependencyCycle(o).passed).toBe(false); });
|
|
it("rejects a card from another run, agent or with one choice", () => {
|
|
expect(gradeDependencyCycle(withCard(row => { row.sourceRunId = "other"; })).passed).toBe(false);
|
|
expect(gradeDependencyCycle(withCard(row => { row.createdByAgentId = "someone"; })).passed).toBe(false);
|
|
expect(gradeDependencyCycle(withCard(row => { row.status = "answered"; })).passed).toBe(false);
|
|
expect(gradeDependencyCycle(withCard(row => { row.payload.questions[0]!.options = row.payload.questions[0]!.options.slice(0, 1); })).passed).toBe(false);
|
|
});
|
|
it("rejects a card that does not name both tasks", () => expect(gradeDependencyCycle(withCard(row => {
|
|
row.title = "Choose order"; row.summary = "That would be a cycle."; row.payload.questions[0]!.prompt = "Which order?";
|
|
row.payload.questions[0]!.options = [{ id: "a", label: "Option A" }, { id: "b", label: "Option B" }];
|
|
})).passed).toBe(false));
|
|
it("rejects a valid card when the original prerequisite was swapped", () => {
|
|
const o = withCard(); o.dependencyTasks[1]!.blockedBy = []; o.dependencyTasks[0]!.blockedBy = [{ id: "design" }];
|
|
expect(gradeDependencyCycle(o).passed).toBe(false);
|
|
});
|