mirror of
https://github.com/paperclipai/paperclip.git
synced 2026-11-13 15:55:46 +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>
79 lines
6.7 KiB
TypeScript
79 lines
6.7 KiB
TypeScript
import { execFileSync } from "node:child_process";
|
|
import { createHash } from "node:crypto";
|
|
import { readFile, writeFile } from "node:fs/promises";
|
|
import path from "node:path";
|
|
import { expect } from "@playwright/test";
|
|
import type { IssueDocument } from "../../../packages/shared/src/types/issue.js";
|
|
import { checks, putDocument } from "./internal-support.js";
|
|
import { privilegedDefinition, privilegedObservation, settlePrivilegedChat } from "./privileged-support.js";
|
|
import { prepareOwnedCli, runOwnedCli, cleanupOwnedCli } from "./privileged-cli-support.js";
|
|
|
|
export function documentExportCommand(reply: string, sourceId: string, sourceIdentifier?: string) {
|
|
const blocks = [...reply.matchAll(/```(?:bash|sh|shell|zsh|console)?\s*\n([\s\S]*?)```/g)].map(match => match[1]!.replace(/\\\r?\n/g, " ").trim());
|
|
const matches = blocks.flatMap(block => block.split(/\r?\n/)).map(line => line.trim().replace(/^\$\s+/, "").match(/^(?:paperclipai|npx(?: -y)? paperclipai|pnpm (?:exec )?paperclipai)\s+issue\s+document:get\s+(\S+)\s+plan\s+--json\s*\|\s*jq\s+(-[rj]+)\s+['"]?\.body['"]?\s*>\s*([A-Za-z0-9][A-Za-z0-9_.-]*)$/)).filter(Boolean);
|
|
if (matches.length !== 1) throw new Error("One supported plan JSON-to-file command is required; arbitrary shell is not admitted");
|
|
const command = matches[0]!;
|
|
// The CLI resolves a task by its UUID or its identifier (such as COM-1).
|
|
const original = command[1]!, id = /^(?:<[A-Z_ -]+>|\$\{?[A-Z_]+\}?)$/.test(original) || original === sourceIdentifier ? sourceId : original;
|
|
if (id !== sourceId || path.basename(command[3]!) !== command[3]!) throw new Error("Document export targets a different task or destination");
|
|
return { argv: ["issue", "document:get", id, "plan", "--json"], filterArgs: [command[2]!, ".body"], filename: command[3]!,
|
|
unresolvedIssueBinding: original === id || original === sourceIdentifier ? undefined : { placeholder: original, fixtureValue: id } };
|
|
}
|
|
export function gradeExportDocumentCli(value: unknown) {
|
|
const c = checks(), o = privilegedObservation(value);
|
|
c.add("saved-document-export-evidence", !!o, "An actual supported CLI read, structured result and owned file bytes are required."); if (!o) return c.result();
|
|
c.add("correct-saved-revision", typeof o.state.documentRevisionId === "string" && o.state.exportRevisionId === o.state.documentRevisionId
|
|
&& o.state.documentUnchanged === true && o.state.exportIssueId === o.state.sourceIssueId,
|
|
"The downloaded structured document is the exact selected saved plan revision on the authorized source task.");
|
|
// jq -r ends its output with one newline; a text file that differs only by
|
|
// that final newline is the same saved document.
|
|
c.add("actual-file-bytes", typeof o.state.actualFileSha256 === "string"
|
|
&& (o.state.actualFileSha256 === o.state.savedBodySha256 || o.state.actualFileSha256 === o.state.savedBodyWithNewlineSha256)
|
|
&& o.state.fileReadBack === true && o.state.fileWithinOwnedDirectory === true,
|
|
"The actual terminal output file is read back and matches the saved document body byte-for-byte.");
|
|
c.add("no-summary-substitute", o.state.cliExecuted === true && typeof o.state.commandSha256 === "string"
|
|
&& o.state.exportStructuredResult === true && o.state.onlyExpectedDocumentReturned === true,
|
|
"A chat summary or unrelated document cannot stand in for the real authorized file download.");
|
|
return c.result();
|
|
}
|
|
const exportPlan = privilegedDefinition("cli-operations.export-document-cli", "privileged-document-cli", {
|
|
async setup(ctx) {
|
|
const operator = await prepareOwnedCli(ctx);
|
|
await ctx.input.api.post(`/api/companies/${ctx.input.fixtures.company.id}/chats/${ctx.input.fixtures.agent.id}`);
|
|
await ctx.refreshIssue(); const issue = ctx.issue(); if (!issue) throw new Error("Owned standing task did not persist");
|
|
const body = `# Website plan\n\nA saved synthetic revision for Acorn.\n\nReference ${ctx.input.nonce}.`;
|
|
const document = await putDocument(ctx, issue.id, body);
|
|
return { ...operator, sourceIssueId: issue.id, documentRevisionId: document.latestRevisionId,
|
|
sourceIdentifier: issue.identifier, savedBodySha256: createHash("sha256").update(body).digest("hex"),
|
|
savedBodyWithNewlineSha256: createHash("sha256").update(`${body}\n`).digest("hex") };
|
|
},
|
|
async act(ctx, state) {
|
|
const reply = await settlePrivilegedChat(ctx), sourceId = String(state.sourceIssueId);
|
|
const command = documentExportCommand(reply.reply, sourceId, typeof state.sourceIdentifier === "string" ? state.sourceIdentifier : undefined);
|
|
const json = await runOwnedCli(ctx, command.argv), document = JSON.parse(json) as IssueDocument;
|
|
// Execute the actual finite jq filter from the returned command; never a shell.
|
|
const bytes = execFileSync("jq", command.filterArgs, { input: json, timeout: 5_000, maxBuffer: 1_048_576 });
|
|
const destination = path.join(String(state.operatorDirectory), command.filename);
|
|
await writeFile(destination, bytes, { flag: "wx", mode: 0o600 });
|
|
const actual = await readFile(destination);
|
|
state.actualFileSha256 = createHash("sha256").update(actual).digest("hex"); state.fileReadBack = true;
|
|
state.fileWithinOwnedDirectory = path.dirname(destination) === state.operatorDirectory;
|
|
state.exportRevisionId = document.latestRevisionId; state.exportIssueId = document.issueId;
|
|
const saved = await ctx.input.api.get<IssueDocument>(`/api/issues/${sourceId}/documents/plan`);
|
|
state.documentUnchanged = saved.latestRevisionId === state.documentRevisionId && createHash("sha256").update(saved.body ?? "").digest("hex") === state.savedBodySha256;
|
|
state.exportStructuredResult = typeof document.body === "string" && typeof document.latestRevisionId === "string";
|
|
state.onlyExpectedDocumentReturned = document.key === "plan" && document.issueId === sourceId;
|
|
state.cliExecuted = true;
|
|
state.commandSha256 = createHash("sha256").update(JSON.stringify(command)).digest("hex");
|
|
state.unresolvedIssueBinding = command.unresolvedIssueBinding;
|
|
await ctx.input.evidence("chat-catalog-cli-plan-file.json", { sourceIssueId: sourceId, revisionId: state.documentRevisionId,
|
|
fileSha256: state.actualFileSha256, expectedSha256: state.savedBodySha256, bytes: actual.length,
|
|
filename: command.filename, commandSha256: state.commandSha256, unresolvedIssueBinding: command.unresolvedIssueBinding });
|
|
}, grade: gradeExportDocumentCli,
|
|
async verifyResult(ctx) { await expect(ctx.input.page.getByTestId("task-chat-composer-input")).toBeVisible(); },
|
|
});
|
|
exportPlan.sourceFiles = [...exportPlan.sourceFiles, "chat-catalog-domains/privileged-cli-support.ts", "../../cli/src/index.ts"];
|
|
exportPlan.oracleVersion = "paperclip.chat-catalog.cli-document.v2";
|
|
exportPlan.cleanup = cleanupOwnedCli;
|
|
export const privilegedDocumentCliCases = [exportPlan];
|