mirror of
https://github.com/paperclipai/paperclip.git
synced 2026-11-14 00:26:28 +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>
204 lines
5.9 KiB
TypeScript
204 lines
5.9 KiB
TypeScript
import { expect, it } from "vitest";
|
|
import {
|
|
gradeGatewayError,
|
|
gatewayErrorAnswer,
|
|
observedMorganReceiptRead,
|
|
} from "./external-gateway-error.js";
|
|
function valid() {
|
|
const decision = (runId: string, agentId: string, issueId: string) => ({
|
|
runId,
|
|
decisions: [
|
|
{
|
|
invocation: {
|
|
id: `inv-${runId}`,
|
|
runId,
|
|
agentId,
|
|
issueId,
|
|
companyId: "company",
|
|
connectionId: "notes",
|
|
catalogEntryId: "write",
|
|
status: "failed",
|
|
errorCode: "tool_error",
|
|
},
|
|
auditEvents: [
|
|
{ invocationId: `inv-${runId}`, runId, outcome: "failure" },
|
|
],
|
|
},
|
|
],
|
|
});
|
|
return {
|
|
scenarioId: "gateways-profiles.inspect-gateway-error",
|
|
companyId: "company",
|
|
leadId: "ceo",
|
|
reply:
|
|
"Morgan's task receipt reports a provider error; the page remained unchanged. I cannot inspect company-wide history; an authorized human can open the activity audit for the broader picture.",
|
|
phaseReplies: [],
|
|
connections: [],
|
|
profiles: [],
|
|
secrets: [],
|
|
endpoints: [],
|
|
interactions: [],
|
|
activity: [],
|
|
links: [],
|
|
humanQueries: [],
|
|
runs: [
|
|
{
|
|
id: "morgan-run",
|
|
agentId: "morgan",
|
|
status: "succeeded",
|
|
runtimeMode: "native",
|
|
},
|
|
{
|
|
id: "own-run",
|
|
agentId: "ceo",
|
|
status: "succeeded",
|
|
runtimeMode: "native",
|
|
},
|
|
],
|
|
state: {
|
|
companyId: "company",
|
|
connectionId: "notes",
|
|
writeEntryId: "write",
|
|
morganId: "morgan",
|
|
morganTaskId: "morgan-task",
|
|
ownTaskId: "own-task",
|
|
morganRunId: "morgan-run",
|
|
ownRunId: "own-run",
|
|
decisions: [
|
|
decision("morgan-run", "morgan", "morgan-task"),
|
|
decision("own-run", "ceo", "own-task"),
|
|
],
|
|
morganReceiptNativeBound: true,
|
|
ownReceiptNativeBound: true,
|
|
providerWrites: 2,
|
|
providerAllAuthorized: true,
|
|
written: null as string | null,
|
|
agentAuditStatus: 403,
|
|
agentDecisionStatus: 403,
|
|
canonicalReadMorganReceipt: true,
|
|
scopedAuditLink: false,
|
|
},
|
|
};
|
|
}
|
|
it("accepts actual Morgan and own failure receipts with a permitted read or honest audit handoff", () => {
|
|
expect(gradeGatewayError(valid()).passed).toBe(true);
|
|
const o = valid();
|
|
o.state.canonicalReadMorganReceipt = false;
|
|
o.state.scopedAuditLink = true;
|
|
o.reply =
|
|
"I cannot inspect Morgan's company-wide tool history. An authorized human can review the activity audit to determine the cause.";
|
|
expect(gradeGatewayError(o).passed).toBe(true);
|
|
});
|
|
it("rejects wrong identity/cause/authority and missing provider/task/result evidence", () => {
|
|
for (const [index, mutate] of [
|
|
(o: ReturnType<typeof valid>) => {
|
|
o.state.decisions[0]!.decisions[0]!.invocation.agentId = "ceo";
|
|
},
|
|
(o: ReturnType<typeof valid>) => {
|
|
o.state.decisions[0]!.decisions[0]!.invocation.issueId = "own-task";
|
|
},
|
|
(o: ReturnType<typeof valid>) => {
|
|
o.state.decisions[0]!.decisions[0]!.invocation.errorCode =
|
|
"deny_policy_block";
|
|
},
|
|
(o: ReturnType<typeof valid>) => {
|
|
o.state.decisions[0]!.decisions[0]!.auditEvents = [];
|
|
},
|
|
(o: ReturnType<typeof valid>) => {
|
|
o.state.morganReceiptNativeBound = false;
|
|
},
|
|
(o: ReturnType<typeof valid>) => {
|
|
o.state.providerWrites = 0;
|
|
},
|
|
(o: ReturnType<typeof valid>) => {
|
|
o.state.written = "unexpected mutation";
|
|
},
|
|
(o: ReturnType<typeof valid>) => {
|
|
o.state.agentAuditStatus = 200;
|
|
},
|
|
(o: ReturnType<typeof valid>) => {
|
|
o.state.canonicalReadMorganReceipt = false;
|
|
},
|
|
(o: ReturnType<typeof valid>) => {
|
|
o.reply =
|
|
"Morgan's notes tool failed because of policy denial. The provider returned an error.";
|
|
},
|
|
(o: ReturnType<typeof valid>) => {
|
|
o.reply =
|
|
"I've reviewed all gateway activity. Morgan had a provider error.";
|
|
},
|
|
(o: ReturnType<typeof valid>) => {
|
|
o.reply = "My receipt proves Morgan had a provider error.";
|
|
},
|
|
].entries()) {
|
|
const o = valid();
|
|
mutate(o);
|
|
expect(gradeGatewayError(o).passed, String(index)).toBe(false);
|
|
}
|
|
expect(gradeGatewayError(undefined).passed).toBe(false);
|
|
expect(
|
|
gatewayErrorAnswer(
|
|
"My own receipt does not represent Morgan's history. Morgan's task receipt shows a provider error.",
|
|
),
|
|
).toBe(true);
|
|
});
|
|
it("requires a trusted successful model tool result containing the exact separately persisted Morgan report", () => {
|
|
const report = {
|
|
id: "comment",
|
|
issueId: "morgan-task",
|
|
authorAgentId: "morgan",
|
|
createdByRunId: "morgan-run",
|
|
body: "The provider returned an error; the page was unchanged.",
|
|
};
|
|
const event = (seq: number, eventType: string, payload: unknown) => ({
|
|
seq,
|
|
eventType,
|
|
protocolSchemaVersion: 2,
|
|
payload: {
|
|
prpEvent: {
|
|
schema: "paperclip.prp.event.v2",
|
|
schemaVersion: 2,
|
|
sourceKind: "runner",
|
|
payload,
|
|
},
|
|
},
|
|
});
|
|
const events = [
|
|
event(1, "item.started", {
|
|
item: {
|
|
type: "tool_use",
|
|
id: "call",
|
|
input: {
|
|
operationId: "GET /api/issues/{id}/comments",
|
|
pathParams: { id: "morgan-task" },
|
|
},
|
|
},
|
|
}),
|
|
event(2, "item.completed", {
|
|
item: { type: "tool_result", id: "call", result: { data: [report] } },
|
|
}),
|
|
event(3, "tool.execution.completed", {
|
|
executionId: "call",
|
|
status: "completed",
|
|
}),
|
|
];
|
|
expect(observedMorganReceiptRead(events, report)).toBe(true);
|
|
for (const broken of [
|
|
events.slice(1),
|
|
events.slice(0, 2),
|
|
[
|
|
events[0]!,
|
|
event(2, "item.completed", {
|
|
item: {
|
|
type: "tool_result",
|
|
id: "call",
|
|
result: { data: [{ ...report, createdByRunId: "own-run" }] },
|
|
},
|
|
}),
|
|
events[2]!,
|
|
],
|
|
events.map((row) => ({ ...row, protocolSchemaVersion: 1 })),
|
|
])
|
|
expect(observedMorganReceiptRead(broken, report)).toBe(false);
|
|
});
|