mirror of
https://github.com/paperclipai/paperclip.git
synced 2026-11-15 10:21:41 +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>
57 lines
6.1 KiB
TypeScript
57 lines
6.1 KiB
TypeScript
import { mkdtemp, mkdir, writeFile, readFile, realpath, rm } from "node:fs/promises";
|
|
import { tmpdir } from "node:os";
|
|
import path from "node:path";
|
|
import { fileURLToPath } from "node:url";
|
|
import { createHash } from "node:crypto";
|
|
import { execFile } from "node:child_process";
|
|
import { promisify } from "node:util";
|
|
import { mirrorPreviewSdk } from "./external-plugin-sdk-mirror.js";
|
|
import type { ChatCatalogContext } from "../chat-catalog-contract.js";
|
|
import type { PaperclipPluginManifestV1 } from "../../../packages/shared/src/types/plugin.js";
|
|
|
|
export function previewFixtureManifest(key: string): PaperclipPluginManifestV1 {
|
|
return { id: key, apiVersion: 1, version: "0.1.0", displayName: "Website Preview Eval", description: "Disposable synthetic website preview fixture", author: "Paperclip evaluation fixture", categories: ["ui", "automation"],
|
|
capabilities: ["api.routes.register", "agent.tools.register", "jobs.schedule", "secrets.read-ref"], entrypoints: { worker: "./dist/worker.js" },
|
|
instanceConfigSchema: { type: "object", properties: { apiKey: { type: "string", title: "Preview credential", format: "secret-ref" } }, required: ["apiKey"], additionalProperties: false },
|
|
jobs: [{ jobKey: "preview-refresh", displayName: "Preview refresh", description: "Synthetic annual refresh; no external service", schedule: "0 0 1 1 *" }],
|
|
apiRoutes: [{ routeKey: "preview", method: "GET", path: "/preview", auth: "board", capability: "api.routes.register", companyResolution: { from: "query", key: "companyId" } }],
|
|
tools: [{ name: "preview", displayName: "Preview website", description: "Read the current synthetic website preview", parametersSchema: { type: "object", properties: {}, additionalProperties: false } }],
|
|
};
|
|
}
|
|
/** CLI scaffolding and package admission run during the selected attempt.
|
|
* The explicit offline calibration exercises scaffolding without admission. */
|
|
export async function createPreviewFixture(ctx: ChatCatalogContext) {
|
|
const temporary = await realpath(await mkdtemp(path.join(tmpdir(), "paperclip-preview-eval-")));
|
|
const packageRoot = path.join(temporary, "website-preview");
|
|
const cli = fileURLToPath(new URL("../../../cli/src/index.ts", import.meta.url));
|
|
const loader = fileURLToPath(new URL("../../../cli/node_modules/tsx/dist/loader.mjs", import.meta.url));
|
|
const sdkUrl = new URL("../../../packages/plugins/sdk/dist/index.js", import.meta.url).href;
|
|
try {
|
|
const mirror = await mirrorPreviewSdk(fileURLToPath(new URL("../../../packages/plugins/sdk/", import.meta.url)), temporary);
|
|
await promisify(execFile)(process.execPath, ["--import", loader, cli, "plugin", "init", "@paperclip-eval/website-preview", "--output", temporary, "--display-name", "Website Preview Eval", "--sdk-path", mirror.sdkPath], { cwd: temporary, timeout: 30_000, maxBuffer: 256_000, env: { HOME: temporary, PATH: process.env.PATH ?? "", NODE_ENV: "test", DO_NOT_TRACK: "1", PAPERCLIP_TELEMETRY_DISABLED: "1" } });
|
|
const key = `paperclip.eval-preview-${ctx.input.nonce.replace(/[^a-z0-9]/gi, "").toLowerCase()}`;
|
|
const manifest = previewFixtureManifest(key);
|
|
const marker = `Synthetic website preview ${ctx.input.nonce}`;
|
|
const manifestSource = `export default ${JSON.stringify(manifest)};\n`;
|
|
// Secret-ref config holds the SDK's { type: "secret_ref", secretId, version } binding.
|
|
// The preview route reports configured only when the bound value actually resolves;
|
|
// the value itself never leaves the worker.
|
|
const workerSource = `import {definePlugin,runWorker} from ${JSON.stringify(sdkUrl)};
|
|
let configured=false,context;
|
|
const bound=value=>!!value&&typeof value==='object'&&value.type==='secret_ref'&&typeof value.secretId==='string'&&value.secretId.length>0;
|
|
const plugin=definePlugin({
|
|
async setup(ctx){context=ctx;ctx.jobs.register('preview-refresh',async()=>{});ctx.tools.register('preview',{displayName:'Preview website',description:'Read the synthetic preview',parametersSchema:{type:'object',properties:{},additionalProperties:false}},async()=>({content:${JSON.stringify(marker)}}));},
|
|
async onConfigChanged(config){configured=bound(config.apiKey);},
|
|
async onValidateConfig(config){return {valid:bound(config.apiKey),message:'Choose the preview credential securely.'};},
|
|
async onHealth(){return {status:configured?'ok':'degraded',message:configured?'Synthetic preview configured':'Preview credential is missing'};},
|
|
async onApiRequest(input){let usable=false;try{const config=await context.config.get(input.companyId);usable=bound(config.apiKey)&&(await context.secrets.resolve(config.apiKey,{companyId:input.companyId,configPath:'apiKey'})).length>0;}catch{usable=false;}return {body:{companyId:input.companyId,marker:${JSON.stringify(marker)},configured:usable}};}
|
|
});export default plugin;runWorker(plugin,import.meta.url);\n`;
|
|
await mkdir(path.join(packageRoot, "dist"), { recursive: true });
|
|
await writeFile(path.join(packageRoot, "package.json"), JSON.stringify({ name: "@paperclip-eval/website-preview", version: "0.1.0", type: "module", private: true, paperclipPlugin: { manifest: "./dist/manifest.js", worker: "./dist/worker.js" } }));
|
|
await writeFile(path.join(packageRoot, "dist/manifest.js"), manifestSource); await writeFile(path.join(packageRoot, "dist/worker.js"), workerSource);
|
|
const digest = (bytes: string | Buffer) => createHash("sha256").update(bytes).digest("hex");
|
|
await ctx.input.evidence("chat-catalog-vetted-preview-package.json", { synthetic: true, cliMode: "supported-source-cli-tsx", cliLoaderSha256: digest(await readFile(loader)), pluginKey: key, sdkMirror: { files: mirror.files, archiveWrites: mirror.archiveWrites, packageBuilds: mirror.packageBuilds, packagePacks: mirror.packagePacks }, manifestSha256: digest(manifestSource), workerSha256: digest(workerSource), cliSha256: digest(await readFile(cli)), sdkEntrySha256: digest(await readFile(fileURLToPath(sdkUrl))) });
|
|
return { key, packageRoot, marker, async close() { await rm(temporary, { recursive: true, force: true }); } };
|
|
} catch { await rm(temporary, { recursive: true, force: true }); throw new Error("Disposable plugin scaffold failed; command output withheld"); }
|
|
}
|