mirror of
https://github.com/paperclipai/paperclip.git
synced 2026-11-15 01:52:55 +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>
64 lines
5.5 KiB
TypeScript
64 lines
5.5 KiB
TypeScript
import { expect, it, afterEach, vi } from "vitest";
|
|
import { mkdtemp, mkdir, writeFile, readFile, rm, symlink, realpath } from "node:fs/promises";
|
|
import path from "node:path";
|
|
import { tmpdir } from "node:os";
|
|
import { spawn } from "node:child_process";
|
|
import type { ChatCatalogContext } from "../chat-catalog-contract.js";
|
|
import { readRuntimeTaskGate, releaseRuntimeWorker, taskBoundRuntimeGate, prepareRuntimeWorkerGate } from "./runtime-busy-worker.js";
|
|
const roots: string[] = [];
|
|
afterEach(async () => { vi.unstubAllEnvs(); for (const root of roots.splice(0)) await rm(root, { recursive: true, force: true }); });
|
|
async function fixture() {
|
|
const root = await mkdtemp(path.join(tmpdir(), "runtime-task-gate-")); roots.push(root); vi.stubEnv("PAPERCLIP_RUNNER_E2E_TEMP_ROOT", root);
|
|
const cwd = path.join(root, "isolated-workspaces", "company", "task"); await mkdir(cwd, { recursive: true });
|
|
const task = { id: "task", companyId: "company", assigneeAgentId: "worker", executionWorkspaceId: "workspace" };
|
|
const workspace = { id: "workspace", companyId: "company", cwd };
|
|
const ctx = { input: { fixtures: { company: { id: "company" } }, api: { get: async (url: string) => url.endsWith("/workspace") ? { workspace } : task } } } as unknown as ChatCatalogContext;
|
|
return { root, cwd, task, workspace, ctx };
|
|
}
|
|
it("a real bounded Node utility creates readiness in actual task cwd and emits only the genuine released brief", async () => {
|
|
const f = await fixture(), gate = taskBoundRuntimeGate("owned-test", 3000);
|
|
expect(await readRuntimeTaskGate(f.ctx, gate, "task", "worker")).toBe("");
|
|
const child = spawn("/bin/sh", ["-c", `exec ${gate.command}`], { cwd: f.cwd, stdio: ["ignore", "pipe", "pipe"] });
|
|
let stdout = ""; child.stdout.on("data", bytes => { stdout += bytes; });
|
|
const exited = new Promise<number | null>(resolve => child.once("exit", resolve));
|
|
try {
|
|
for (let i = 0; i < 50 && await readRuntimeTaskGate(f.ctx, gate, "task", "worker") !== "waiting"; i++) await new Promise(resolve => setTimeout(resolve, 20));
|
|
expect(await readFile(gate.ready, "utf8")).toBe("waiting"); expect(child.exitCode).toBeNull(); expect(stdout).toBe("");
|
|
await writeFile(gate.gate, "Actual human brief: preserve existing website.");
|
|
expect(await exited).toBe(0); expect(stdout.trim()).toBe("Actual human brief: preserve existing website.");
|
|
expect(gate.gate.startsWith(await realpath(f.cwd) + path.sep)).toBe(true);
|
|
} finally { if (child.exitCode === null) { child.kill("SIGTERM"); await exited; } }
|
|
});
|
|
it("rejects another company, assignee, active binding or changed workspace before reading/releasing", async () => {
|
|
const f = await fixture(), gate = taskBoundRuntimeGate("scope-test");
|
|
f.task.assigneeAgentId = "other"; await expect(readRuntimeTaskGate(f.ctx, gate, "task", "worker")).rejects.toThrow("assignee");
|
|
f.task.assigneeAgentId = "worker"; f.workspace.companyId = "foreign"; await expect(readRuntimeTaskGate(f.ctx, gate, "task", "worker")).rejects.toThrow("company");
|
|
f.workspace.companyId = "company"; f.task.executionWorkspaceId = "other"; await expect(readRuntimeTaskGate(f.ctx, gate, "task", "worker")).rejects.toThrow("binding");
|
|
f.task.executionWorkspaceId = "workspace"; await readRuntimeTaskGate(f.ctx, gate, "task", "worker");
|
|
f.workspace.id = "new"; f.task.executionWorkspaceId = "new";
|
|
await expect(readRuntimeTaskGate(f.ctx, gate, "task", "worker")).rejects.toThrow("changed after admission");
|
|
});
|
|
it("rejects escaping workspace and readiness symlinks instead of treating foreign bytes as busy evidence", async () => {
|
|
const f = await fixture(), gate = taskBoundRuntimeGate("symlink-test");
|
|
await readRuntimeTaskGate(f.ctx, gate, "task", "worker");
|
|
await writeFile(path.join(f.root, "outside-ready"), "waiting"); await symlink(path.join(f.root, "outside-ready"), gate.ready);
|
|
await expect(readRuntimeTaskGate(f.ctx, gate, "task", "worker")).rejects.toThrow("readiness path escapes");
|
|
f.workspace.cwd = tmpdir();
|
|
await expect(readRuntimeTaskGate(f.ctx, taskBoundRuntimeGate("outside"), "task", "worker")).rejects.toThrow("owned attempt root");
|
|
});
|
|
it("retains neutral cleanup release and refuses timeout widening", async () => {
|
|
const f = await fixture(), gate = taskBoundRuntimeGate("cleanup"); await readRuntimeTaskGate(f.ctx, gate, "task", "worker");
|
|
await releaseRuntimeWorker({ gate });
|
|
expect(await readFile(gate.gate, "utf8")).toBe("The human ended this disposable fixture. Preserve any saved work and finish normally.");
|
|
expect(() => taskBoundRuntimeGate("too-long", 300001)).toThrow("bounded");
|
|
});
|
|
it("funds source and worker before task-bound history, without exposing an agent-home script or admitting work", async () => {
|
|
const f = await fixture(); vi.stubEnv("PAPERCLIP_HOME", path.join(f.root, "paperclip-home")); vi.stubEnv("PAPERCLIP_INSTANCE_ID", "owned-test");
|
|
const funded: string[] = [];
|
|
const ctx = { input: { nonce: "history", execution: { environment: { id: "local" } }, fixtures: { company: { id: "company" }, agent: { id: "source-agent" } },
|
|
api: { post: async (_url: string, body: { scopeId: string; unpricedUsagePolicy: string }) => { expect(body.unpricedUsagePolicy).toBe("allow"); funded.push(body.scopeId); } } }, allRuns: async () => [] } as unknown as ChatCatalogContext;
|
|
const prepared = await prepareRuntimeWorkerGate(ctx, { worker: { id: "worker" }, taskBound: true });
|
|
expect(funded).toEqual(["company", "worker", "source-agent"]);
|
|
expect(prepared.gate.scriptPath).toBe(""); expect(prepared.gate.command).not.toContain(f.root);
|
|
});
|