mirror of
https://github.com/paperclipai/paperclip.git
synced 2026-11-15 18:56:05 +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>
47 lines
4.7 KiB
TypeScript
47 lines
4.7 KiB
TypeScript
import { execFile } from "node:child_process";
|
|
import { promisify } from "node:util";
|
|
import { mkdtemp, writeFile, rm } from "node:fs/promises";
|
|
import { tmpdir } from "node:os";
|
|
import path from "node:path";
|
|
import { fileURLToPath } from "node:url";
|
|
import type { Environment } from "../../../packages/shared/src/types/environment.js";
|
|
import type { Cookie } from "@playwright/test";
|
|
export function matchesClaudeFixtureTarget(input: { company: { id: string }; expectedCompanyId: string; fixtureEnvironmentId: string; detail: Pick<Environment, "id" | "driver" | "status">; listed: Array<Pick<Environment, "id" | "driver" | "status">> }) {
|
|
// Environments are instance resources, not company rows. The authenticated
|
|
// company read + public inventory authority and exact attempt fixture bind it.
|
|
const { detail } = input;
|
|
return input.company.id === input.expectedCompanyId && detail.id === input.fixtureEnvironmentId && detail.driver === "local" && detail.status === "active"
|
|
&& input.listed.some(row => row.id === detail.id && row.driver === detail.driver && row.status === detail.status);
|
|
}
|
|
export interface ClaudePrivateReceipt {
|
|
processIsolation: true; testRunnerAttached: false; apiStepRecorder: false;
|
|
operation: "check-harness" | "prepare-account" | "validate-saved"; companyId: string; environmentId: string; connectionId?: string; grantId?: string; agentId?: string;
|
|
probeStatus: "pass" | "warn" | "fail"; probeCheckCodes: string[];
|
|
}
|
|
export function validateClaudePrivateReceipt(value: unknown, expected: Pick<ClaudePrivateReceipt, "operation" | "companyId" | "environmentId">): ClaudePrivateReceipt {
|
|
const r = value as ClaudePrivateReceipt;
|
|
if (!r || r.processIsolation !== true || r.testRunnerAttached !== false || r.apiStepRecorder !== false || r.operation !== expected.operation || r.companyId !== expected.companyId || r.environmentId !== expected.environmentId
|
|
|| Object.keys(r).some(key => !["processIsolation", "testRunnerAttached", "apiStepRecorder", "operation", "companyId", "environmentId", "connectionId", "grantId", "agentId", "probeStatus", "probeCheckCodes"].includes(key))
|
|
|| ![r.companyId, r.environmentId, ...(r.operation === "check-harness" ? [] : [r.connectionId, r.grantId]), ...(r.agentId ? [r.agentId] : [])].every(id => typeof id === "string" && /^[a-f0-9-]{36}$/i.test(id))
|
|
|| !["pass", "warn", "fail"].includes(r.probeStatus) || !Array.isArray(r.probeCheckCodes) || r.probeCheckCodes.length > 100 || !r.probeCheckCodes.every(code => /^[a-z0-9_.-]{1,100}$/i.test(code))) throw new Error("Invalid closed Claude qualification receipt; private diagnostics withheld");
|
|
return r;
|
|
}
|
|
export type ClaudePrivateRequest = { operation: "check-harness"; harnessExecutable: string } | { operation: "prepare-account"; credential: string; harnessExecutable: string; model: string } | { operation: "validate-saved"; agentId: string; connectionId: string; grantId: string };
|
|
/** Only the selected per-attempt key/session crosses the private0600 file.
|
|
* No Test request API, browser arguments, argv, env or raw child diagnostic gets it. */
|
|
export async function privateClaudeOperation(input: ClaudePrivateRequest & { companyId: string; environmentId: string; baseURL: string; cookies: Cookie[] }, scanSecret = input.operation === "prepare-account" ? input.credential : undefined) {
|
|
if (!scanSecret || scanSecret.length < 12) throw new Error("Claude private operation requires the actual selected credential for private receipt scanning");
|
|
const directory = await mkdtemp(path.join(tmpdir(), "paperclip-claude-private-"));
|
|
try {
|
|
const request = path.join(directory, "request.json"); await writeFile(request, JSON.stringify(input), { mode: 0o600 });
|
|
const loader = fileURLToPath(new URL("../../../server/node_modules/tsx/dist/loader.mjs", import.meta.url)), worker = fileURLToPath(new URL("./runtime-claude-private-worker.ts", import.meta.url));
|
|
let output: string;
|
|
try { output = (await promisify(execFile)(process.execPath, ["--import", loader, worker, request], { timeout: 180000, maxBuffer: 32768,
|
|
env: { PATH: process.env.PATH ?? "", HOME: directory, TMPDIR: directory, NODE_ENV: "test", DO_NOT_TRACK: "1" } })).stdout; }
|
|
catch { throw new Error("Selected Claude public setup/validation failed; private request and child diagnostics withheld"); }
|
|
if ([scanSecret, encodeURIComponent(scanSecret), Buffer.from(scanSecret).toString("base64")].some(value => output.includes(value))) throw new Error("Claude private child exceeded its secret-free receipt boundary; output withheld");
|
|
let result: unknown; try { result = JSON.parse(output); } catch { throw new Error("Claude private child returned no closed qualification receipt"); }
|
|
return validateClaudePrivateReceipt(result, input);
|
|
} finally { await rm(directory, { recursive: true, force: true }); }
|
|
}
|