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>
70 lines
6.0 KiB
TypeScript
70 lines
6.0 KiB
TypeScript
import { cleanupIdentityProbes } from "./identity-support.js";
|
|
import { expect } from "@playwright/test";
|
|
import type { ChatCatalogCase, ChatCatalogContext } from "../chat-catalog-contract.js";
|
|
import { observeInternal, checks, type Entity, type InternalObservation } from "./internal-support.js";
|
|
|
|
export interface ExternalObservation extends InternalObservation {
|
|
phaseReplies: Array<{ runId: string; body: string; commentId?: string }>;
|
|
connections: Entity[]; profiles: Entity[]; secrets: Entity[]; endpoints: Entity[];
|
|
interactions: Array<Entity & { kind: string; payload?: Record<string, unknown>; resolvedByUserId?: string; resolvedByAgentId?: string }>;
|
|
}
|
|
export const root = (ctx: ChatCatalogContext) => `/api/companies/${ctx.input.fixtures.company.id}`;
|
|
export async function observeExternal(ctx: ChatCatalogContext, state: Record<string, unknown>): Promise<ExternalObservation> {
|
|
const o = await observeInternal(ctx, state);
|
|
const [connections, profiles, secrets, endpoints, interactions] = await Promise.all([
|
|
ctx.input.api.get<{ connections: Entity[] }>(`${root(ctx)}/tools/connections`), ctx.input.api.get<{ profiles: Entity[] }>(`${root(ctx)}/tools/profiles`),
|
|
ctx.input.api.get<Entity[]>(`${root(ctx)}/secrets`), ctx.input.api.get<Entity[]>(`${root(ctx)}/chat-endpoints`),
|
|
ctx.issue() ? ctx.input.api.get<ExternalObservation["interactions"]>(`/api/issues/${ctx.issue()!.id}/interactions`) : Promise.resolve([]),
|
|
]);
|
|
const comments = ctx.issue() ? await ctx.input.api.get<Entity[]>(`/api/issues/${ctx.issue()!.id}/comments?order=asc`) : [];
|
|
const phaseReplies = comments.filter(row => row.authorAgentId === ctx.input.fixtures.agent.id && o.runs.some(run => run.id === row.createdByRunId))
|
|
.map(row => ({ runId: row.createdByRunId!, commentId: row.id, body: row.body ?? "" }));
|
|
return { ...o, phaseReplies, connections: connections.connections, profiles: profiles.profiles, secrets, endpoints, interactions };
|
|
}
|
|
export function externalObservation(value: unknown): ExternalObservation | undefined {
|
|
const o = value as ExternalObservation | undefined;
|
|
return o && typeof o.scenarioId === "string" && typeof o.reply === "string" && typeof o.companyId === "string" && !!o.state
|
|
&& [o.phaseReplies, o.connections, o.profiles, o.secrets, o.endpoints, o.interactions, o.activity, o.links, o.runs, o.humanQueries].every(Array.isArray) ? o : undefined;
|
|
}
|
|
export function definition(id: string, setup: ChatCatalogCase["setup"], grade: ChatCatalogCase["grade"], extra: Partial<ChatCatalogCase> = {}): ChatCatalogCase {
|
|
const result: ChatCatalogCase = { scenarioId: id, setup, grade, observe: observeExternal,
|
|
oracleVersion: "paperclip.chat-catalog.external.v1", expectedRunCount: 1, turnTimeoutMs: 120_000, attemptTimeoutMs: 300_000,
|
|
sourceFiles: ["chat-catalog-domains/external-support.ts", "chat-catalog-domains/internal-support.ts", `chat-catalog-domains/external-${id.split(".")[0]}.ts`, `chat-scenarios/${id.split(".")[0]}.yaml`],
|
|
async verifyResult(ctx, state, value) {
|
|
const o = externalObservation(value); if (!o) throw new Error("Complete external result evidence required");
|
|
const paths = [state.resultPath, ...(Array.isArray(state.resultPaths) ? state.resultPaths : [])].filter((path): path is string => typeof path === "string");
|
|
if (!paths.length) throw new Error("Case must declare its public result surface");
|
|
const href = o.links.find(link => {
|
|
try { const url = new URL(link, o.origin); return url.origin === o.origin && paths.includes(url.pathname); } catch { return false; }
|
|
});
|
|
if (!href) throw new Error("The reply did not render the intended public management/result link");
|
|
const source = ctx.input.page.url();
|
|
await ctx.input.page.locator(`[id^="comment-"] a[href=${JSON.stringify(href)}]`).first().click();
|
|
await ctx.input.page.waitForURL(url => paths.includes(url.pathname));
|
|
await expect(ctx.input.page.getByRole("main")).toBeVisible();
|
|
if (typeof state.resultLabel === "string") await expect(ctx.input.page.getByText(state.resultLabel, { exact: true }).first()).toBeVisible();
|
|
await ctx.input.evidence("chat-catalog-external-result.json", { scenarioId: o.scenarioId, pathname: new URL(ctx.input.page.url()).pathname, label: state.resultLabel });
|
|
await ctx.input.page.goto(source, { waitUntil: "domcontentloaded" });
|
|
await expect(ctx.input.page.getByTestId("task-chat-composer-input")).toBeVisible();
|
|
}, ...extra };
|
|
return { ...result, sourceFiles: [...new Set([...result.sourceFiles, "chat-catalog-domains/identity-support.ts"])],
|
|
async setup(ctx) { try { return await result.setup(ctx); } catch (error) { await cleanupIdentityProbes(ctx); throw error; } },
|
|
async cleanup(ctx, state) { try { await result.cleanup?.(ctx, state); } finally { await cleanupIdentityProbes(ctx); } },
|
|
};
|
|
}
|
|
/** True when the reply rendered a same-origin link to one of the case's setup-chosen pages. */
|
|
export function repliedWithPageLink(o: { links: string[]; origin: string }, path: unknown): boolean {
|
|
return typeof path === "string" && o.links.some(href => { try { const url = new URL(href, o.origin); return url.origin === o.origin && url.pathname === path; } catch { return false; } });
|
|
}
|
|
export function sameRows(before: unknown, after: unknown): boolean { return JSON.stringify(before) === JSON.stringify(after); }
|
|
export { checks };
|
|
|
|
/** Bind visible result verification to the exact persisted model comment. */
|
|
export async function verifyRenderedReply(ctx: ChatCatalogContext, value: unknown, excludeRunId?: string) {
|
|
const o = externalObservation(value); if (!o) throw new Error("Missing rendered reply observation");
|
|
const reply = o.phaseReplies.filter(row => row.runId !== excludeRunId).at(-1);
|
|
if (!reply?.commentId || !reply.body.trim()) throw new Error("No persisted result comment for this phase");
|
|
await expect(ctx.input.page.locator(`[id=${JSON.stringify(`comment-${reply.commentId}`)}]`)).toBeVisible();
|
|
await ctx.input.evidence("chat-catalog-external-visible-reply.json", { scenarioId: o.scenarioId, commentId: reply.commentId, runId: reply.runId });
|
|
}
|