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>
57 lines
4.6 KiB
TypeScript
57 lines
4.6 KiB
TypeScript
import { it, expect } from "vitest";
|
|
import { gradeAgentKey, containsPrivateCredential } from "./runtime-agent-keys.js";
|
|
function valid() {
|
|
const worker = { id: "morgan", companyId: "company", name: "Morgan", permissions: {}, adapterConfig: {} };
|
|
const key = { id: "key", name: "local integration", revokedAt: null as string | null };
|
|
return { scenarioId: "agents.agent-api-key", origin: "http://fixture.invalid", companyId: "company", leadId: "ceo",
|
|
state: { worker, agentCreateStatus: 403, keysBeforeHuman: 0, humanCreated: true, createdKey: key, createdKeyAgentId: "morgan", createdKeyCompanyId: "company",
|
|
privateBrowserProcessIsolated: true, privateBrowserTestRunnerAttached: false, privateBrowserApiStepRecorder: false, tokenShownOnce: true, tokenAbsentAfterReload: true, privateScanComplete: true, artifactScanComplete: true, runEvidenceScanComplete: true, tokenLeaked: false,
|
|
revokedCredentialStatus: 401, finalKeys: [{ ...key, revokedAt: "2026-10-26T00:00:00Z" }] },
|
|
agents: [structuredClone(worker)], tasks: [], links: ["/RUN/agents/morgan/api-keys"], reply: "Use Morgan's profile API key controls; the board can create the credential securely.",
|
|
activity: ["agent.key_created", "agent.key_revoked"].map(action => ({ id: action, action, entityId: "morgan", actorType: "user", details: { keyId: "key" } })),
|
|
runs: [], humanQueries: [] };
|
|
}
|
|
it("accepts secure one-time human creation and actual credential revocation", () => expect(gradeAgentKey(valid()).passed).toBe(true));
|
|
it("rejects a mere fresh context still attached to Test API-step recording", () => { const o = valid(); o.state.privateBrowserApiStepRecorder = true; expect(gradeAgentKey(o).passed).toBe(false); });
|
|
it("accepts natural secure human-key guidance", () => { const o = valid(); o.reply = "Open Morgan's profile API key controls and create the local integration credential in your board session. Keep the one-time value private."; expect(gradeAgentKey(o).passed).toBe(true); });
|
|
it("grades the linked controls, not the reply wording", () => {
|
|
const o = valid(); o.reply = "Only a board member can do that. Open [Morgan's keys page](/RUN/agents/morgan/api-keys) and choose Create; copy the value right away."; expect(gradeAgentKey(o).passed).toBe(true);
|
|
o.reply = "That needs a board session. Use this page: [Morgan](/RUN/agents/morgan/keys)."; o.links = ["/RUN/agents/morgan/keys"]; expect(gradeAgentKey(o).passed).toBe(true);
|
|
});
|
|
it("rejects a bare profile link because the profile navigation has no API keys item", () => {
|
|
const o = valid(); o.links = ["/RUN/agents/morgan"]; expect(gradeAgentKey(o).passed).toBe(false);
|
|
});
|
|
it("rejects another agent's key page and a reply without any link", () => {
|
|
const o = valid(); o.links = ["/RUN/agents/other/api-keys"]; expect(gradeAgentKey(o).passed).toBe(false);
|
|
o.links = []; expect(gradeAgentKey(o).passed).toBe(false);
|
|
});
|
|
it("rejects an agent-created key even when the reply links the controls", () => {
|
|
const o = valid(); o.state.keysBeforeHuman = 1; expect(gradeAgentKey(o).passed).toBe(false);
|
|
});
|
|
it("rejects missing human credential evidence", () => expect(gradeAgentKey(undefined).passed).toBe(false));
|
|
it("rejects a key that authenticates a different agent", () => {
|
|
const o = valid(); o.state.createdKeyAgentId = "other"; expect(gradeAgentKey(o).passed).toBe(false);
|
|
});
|
|
it("rejects revocation metadata while the credential still works", () => {
|
|
const o = valid(); o.state.revokedCredentialStatus = 200; expect(gradeAgentKey(o).passed).toBe(false);
|
|
});
|
|
it("rejects credentials copied into conversation artifacts", () => {
|
|
const o = valid(); o.state.tokenLeaked = true; expect(gradeAgentKey(o).passed).toBe(false);
|
|
});
|
|
|
|
it("rejects incomplete generated artifact scans", () => {
|
|
const o = valid(); o.state.artifactScanComplete = false; expect(gradeAgentKey(o).passed).toBe(false);
|
|
});
|
|
it("rejects incomplete run log or event scans", () => {
|
|
const o = valid(); o.state.runEvidenceScanComplete = false; expect(gradeAgentKey(o).passed).toBe(false);
|
|
});
|
|
it("privately detects a synthetic credential in generated artifact bytes", () => {
|
|
const synthetic = "pcp_obviously_synthetic_calibration_credential";
|
|
expect(containsPrivateCredential(["safe reply", { artifact: `prefix ${synthetic} suffix` }], synthetic)).toBe(true);
|
|
expect(containsPrivateCredential(["safe reply", { artifact: "no credential" }], synthetic)).toBe(false);
|
|
});
|
|
it("privately detects a base64 encoded synthetic credential in run evidence", () => {
|
|
const synthetic = "pcp_obviously_synthetic_calibration_credential";
|
|
expect(containsPrivateCredential([{ event: Buffer.from(synthetic).toString("base64") }], synthetic)).toBe(true);
|
|
});
|