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>
58 lines
5.5 KiB
TypeScript
58 lines
5.5 KiB
TypeScript
import { mkdtemp, mkdir, writeFile, readFile, rm } from "node:fs/promises";
|
|
import path from "node:path";
|
|
import { tmpdir } from "node:os";
|
|
import type { SkillSourceRefreshResult } from "../../../packages/shared/src/types/skill-source.js";
|
|
import { accessibilityUpdatePins } from "./integrator-skill-pins.js";
|
|
import { createHash } from "node:crypto";
|
|
import type { ChatCatalogContext } from "../chat-catalog-contract.js";
|
|
import type { CompanySkill, CompanySkillDetail } from "../../../packages/shared/src/types/company-skill.js";
|
|
import { websiteChecklist } from "./integrator-skills.js";
|
|
export const sourceSkillMarkdown = `---\nname: accessibility-review\ndescription: Review a synthetic website for accessibility and report evidence.\n---\n\n# Website accessibility review\n\n${websiteChecklist.map(item => `- ${item}`).join("\n")}\n\nFor each check, record the page and observed evidence; report failures without claiming every page was tested.\n`;
|
|
export const mobileChecks = ["Check touch targets are at least 44 by 44 pixels.", "Repeat the review in both portrait and landscape orientation.", "Verify the checkout interaction on mobile Safari."];
|
|
const sources = new WeakMap<ChatCatalogContext, Awaited<ReturnType<typeof openSource>>>();
|
|
async function openSource() {
|
|
const directory = await mkdtemp(path.join(tmpdir(), "paperclip-vetted-accessibility-"));
|
|
const skillDirectory = path.join(directory, "vetted-candidates", "accessibility-review");
|
|
await mkdir(skillDirectory, { recursive: true });
|
|
const file = path.join(skillDirectory, "SKILL.md");
|
|
await writeFile(file, sourceSkillMarkdown, { mode: 0o600 });
|
|
return { url: skillDirectory, workspaceRoot: directory, requests: [] as Array<{ method: string; path: string; at: number; sha256: string }>,
|
|
async close() { await rm(directory, { recursive: true, force: true }); } };
|
|
}
|
|
export async function openVettedSkillSource(ctx: ChatCatalogContext) {
|
|
if (sources.has(ctx)) throw new Error("Only one owned synthetic skill source is allowed per attempt");
|
|
const source = await openSource(); sources.set(ctx, source); return source;
|
|
}
|
|
/** A source-read receipt is derived only after the evaluated public import,
|
|
* never prepopulated by setup. The actual source bytes, saved revision and
|
|
* agent-attributed import must agree before the caller records it. */
|
|
export async function recordVettedSourceImport(ctx: ChatCatalogContext, source: string, markdown: string) {
|
|
const owned = sources.get(ctx); if (!owned || owned.url !== source) return;
|
|
const actual = await readFile(path.join(owned.url, "SKILL.md"), "utf8");
|
|
if (actual !== sourceSkillMarkdown || actual !== markdown) throw new Error("Vetted source bytes drifted from actual imported revision");
|
|
if (!owned.requests.length) owned.requests.push({ method: "public-local-import", path: "SKILL.md", at: Date.now(), sha256: createHash("sha256").update(actual).digest("hex") });
|
|
}
|
|
export function skillSourceRequests(ctx: ChatCatalogContext) { return sources.get(ctx)?.requests.map(row => ({ ...row })) ?? []; }
|
|
export async function closeVettedSkillSource(ctx: ChatCatalogContext) { const source = sources.get(ctx); if (source) { await source.close(); sources.delete(ctx); } }
|
|
/** Genuine public immutable upstream; local filesystem mode bits are never
|
|
* presented as product-level read-only provenance. */
|
|
export async function importReadOnlySkill(ctx: ChatCatalogContext): Promise<CompanySkillDetail> {
|
|
const pins = accessibilityUpdatePins;
|
|
const response = await fetch(`https://raw.githubusercontent.com/KreerC/ACCESSIBILITY.md/${pins.oldCommit}/${pins.path}`, { redirect: "error", signal: AbortSignal.timeout(30_000) });
|
|
if (!response.ok) { await response.body?.cancel(); throw new Error("Pinned public read-only checklist source unavailable"); }
|
|
const markdown = await response.text();
|
|
if (createHash("sha256").update(markdown).digest("hex") !== pins.oldSha256) throw new Error("Pinned public checklist content changed");
|
|
const root = `/api/companies/${ctx.input.fixtures.company.id}`;
|
|
const imported = await ctx.input.api.post<SkillSourceRefreshResult>(`${root}/skill-sources`, { repositoryUrl: pins.repositoryURL, trackingRef: pins.oldCommit, commitSha: pins.oldCommit, selectedPaths: [pins.path], excludedFolders: [] });
|
|
if (imported.source.lastScanCommit !== pins.oldCommit || imported.imported.length !== 1 || !imported.source.entries.some(entry => entry.path === pins.path && entry.skillId === imported.imported[0]?.id)) throw new Error("Pinned source must produce one actual read-only checklist");
|
|
const skill = await ctx.input.api.get<CompanySkillDetail>(`${root}/skills/${imported.imported[0]!.id}`);
|
|
if (skill.sourceType !== "github" || skill.sourceRef !== pins.oldCommit || skill.editable || skill.markdown !== markdown || !skill.currentVersionId) throw new Error("Actual read-only checklist identity/content/version was not established");
|
|
return skill;
|
|
}
|
|
export async function editableLocalNotes(ctx: ChatCatalogContext, upstream: CompanySkillDetail) {
|
|
const root = `/api/companies/${ctx.input.fixtures.company.id}/skills`;
|
|
const result = await ctx.input.api.post<{ skill: CompanySkill }>(`${root}/${upstream.id}/fork`, { name: "Accessibility local notes", slug: "accessibility-local-notes" });
|
|
await ctx.input.api.patch(`${root}/${result.skill.id}/files`, { path: "SKILL.md", content: `${result.skill.markdown.replace(/^name:[^\n]*$/m, "name: accessibility-local-notes")}\n\n## Local review notes\nKeep ORCHID customer escalation notes unchanged.\n`, expectedVersionId: result.skill.currentVersionId });
|
|
return ctx.input.api.get<CompanySkillDetail>(`${root}/${result.skill.id}`);
|
|
}
|