mirror of
https://github.com/paperclipai/paperclip.git
synced 2026-11-14 17:22:00 +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>
65 lines
5.8 KiB
TypeScript
65 lines
5.8 KiB
TypeScript
import { execFile } from "node:child_process";
|
|
import { promisify } from "node:util";
|
|
import { createHash } from "node:crypto";
|
|
import { readFile, realpath, stat, readdir, lstat, readlink } from "node:fs/promises";
|
|
import path from "node:path";
|
|
import type { ChatCatalogContext } from "../chat-catalog-contract.js";
|
|
import { installedReleaseLaunch } from "../installed-release.js";
|
|
const execute = promisify(execFile);
|
|
export interface InstalledOperatorReceipt { schemaVersion: 1; disposable: true; cli: string; cliSha256: string; server: string; serverSha256: string;
|
|
installationTreeSha256: string; serverPid: number; version: string; targetVersion?: string; targetIntegrity?: string; piCredentialEnv?: string }
|
|
export function validInstalledReceipt(r: InstalledOperatorReceipt) {
|
|
return r.schemaVersion === 1 && r.disposable === true && Number.isSafeInteger(r.serverPid) && r.serverPid > 1
|
|
&& /^[0-9a-f]{64}$/.test(r.installationTreeSha256) && /^[0-9a-f]{64}$/.test(r.cliSha256) && /^[0-9a-f]{64}$/.test(r.serverSha256)
|
|
&& /^\d+\.\d+\.\d+(?:[-+][A-Za-z0-9.-]+)?$/.test(r.version)
|
|
&& (!r.targetVersion || /^\d+\.\d+\.\d+(?:[-+][A-Za-z0-9.-]+)?$/.test(r.targetVersion))
|
|
&& (!r.targetIntegrity || /^sha512-[A-Za-z0-9+/]+=*$/.test(r.targetIntegrity))
|
|
&& (!r.piCredentialEnv || /^[A-Z][A-Z0-9_]+$/.test(r.piCredentialEnv));
|
|
}
|
|
/** Receipt identifies the *actual source server*, not another healthy fixture. */
|
|
export async function ownedInstalledTarget(ctx: ChatCatalogContext) {
|
|
const ref = process.env.PAPERCLIP_CHAT_OPERATOR_INSTALL_RECEIPT, temp = process.env.PAPERCLIP_RUNNER_E2E_TEMP_ROOT;
|
|
if (!ref || !temp) throw new Error("Requires actual owned published source target and PAPERCLIP_CHAT_OPERATOR_INSTALL_RECEIPT (private mode0600; CLI/server hashes, version and live PID)");
|
|
const root = await realpath(temp), receiptFile = await realpath(ref), info = await stat(receiptFile);
|
|
if (!receiptFile.startsWith(root + path.sep) || !info.isFile() || info.mode & 0o077) throw new Error("Installed operator receipt must be private and inside the owned attempt");
|
|
const r = JSON.parse(await readFile(receiptFile, "utf8")) as InstalledOperatorReceipt;
|
|
if (!validInstalledReceipt(r)) throw new Error("Installed operator receipt lacks exact artifact identities");
|
|
const cli = await realpath(r.cli), server = await realpath(r.server);
|
|
if (![cli, server].every(file => file.startsWith(root + path.sep))) throw new Error("Installed operation refuses artifacts outside its owned attempt");
|
|
const launch = installedReleaseLaunch(cli);
|
|
const serverManifest = JSON.parse(await readFile(path.resolve(server, "../../package.json"), "utf8"));
|
|
if (launch.version !== r.version || serverManifest.name !== "@paperclipai/server" || serverManifest.version !== r.version
|
|
|| createHash("sha256").update(await readFile(cli)).digest("hex") !== r.cliSha256
|
|
|| createHash("sha256").update(await readFile(server)).digest("hex") !== r.serverSha256) throw new Error("Installed artifact identity differs from receipt");
|
|
const row = (await execute("ps", ["-p", String(r.serverPid), "-o", "uid=", "-o", "command="], { timeout: 5000 })).stdout.trim();
|
|
const match = row.match(/^(\d+)\s+([\s\S]+)$/);
|
|
if (Number(match?.[1]) !== process.getuid?.() || !match?.[2]?.split(/\s+/).includes(cli)) throw new Error("Receipt does not identify the current user's exact installed CLI process");
|
|
const health = await ctx.input.api.get<{ version?: string }>("/api/health");
|
|
if (health.version !== r.version) throw new Error("Actual source HTTP version differs from the pinned installed target");
|
|
const prefix = path.resolve(launch.cwd, "../../..");
|
|
if (!prefix.startsWith(root + path.sep) || !server.startsWith(prefix + path.sep)) throw new Error("Installed artifacts do not share the exact owned npm prefix");
|
|
const entries: Array<{ file: string; sha256?: string; link?: string }> = []; let totalBytes = 0;
|
|
async function inventory(current: string): Promise<void> {
|
|
for (const name of (await readdir(current)).sort()) {
|
|
const file = path.join(current, name), info = await lstat(file);
|
|
if (entries.length > 200000 || totalBytes > 2_147_483_648) throw new Error("Installed artifact inventory exceeds the bounded proof scope");
|
|
if (info.isSymbolicLink()) { const resolved = await realpath(file); if (!resolved.startsWith(prefix + path.sep)) throw new Error("Installed dependency escapes its owned prefix"); entries.push({ file: path.relative(prefix, file), link: await readlink(file) }); }
|
|
else if (info.isDirectory()) await inventory(file);
|
|
else if (info.isFile()) { totalBytes += info.size; entries.push({ file: path.relative(prefix, file), sha256: createHash("sha256").update(await readFile(file)).digest("hex") }); }
|
|
else throw new Error("Installed artifact inventory contains a nonregular object");
|
|
}
|
|
}
|
|
await inventory(prefix);
|
|
const treeSha256 = createHash("sha256").update(JSON.stringify(entries)).digest("hex");
|
|
if (treeSha256 !== r.installationTreeSha256) throw new Error("Actual installed source dependency tree differs from its pinned artifact receipt");
|
|
const config = process.env.PAPERCLIP_CONFIG;
|
|
if (!config || !(await realpath(config)).startsWith(root + path.sep)) throw new Error("Installed operation requires the owned source config");
|
|
const environment: NodeJS.ProcessEnv = { PATH: process.env.PATH, HOME: root, PAPERCLIP_HOME: root, PAPERCLIP_CONFIG: config,
|
|
PAPERCLIP_TELEMETRY_DISABLED: "1", PAPERCLIP_NO_BROWSER: "1", NO_COLOR: "1", NPM_CONFIG_PREFIX: path.resolve(launch.cwd, "../../..") };
|
|
return { receipt: r, cli, server, root, treeSha256, environment,
|
|
async command(args: string[], timeout = 180_000) {
|
|
try { return (await execute(process.execPath, [cli, ...args], { cwd: launch.cwd, env: environment, timeout, maxBuffer: 1_048_576 })).stdout; }
|
|
catch { throw new Error("Owned installed operator command failed; raw subprocess output withheld"); }
|
|
} };
|
|
}
|