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 eval catalog (`tests/runner-e2e`) runs real chat journeys against a real server and grades the saved state, the run's tool events and the rendered reply > - The full-cohort run on master `f4337bb5fb` had 12 failures that come from the harness, not from the product or the model > - Six plugin and secrets cells stopped in 7–9 s, before the server started, because the SDK public source pins were stale > - Four more cells failed on harness timing, a first-visit prompt, or a read receipt that ignored a dedicated tool > - This pull request fixes those four harness defects and adds a cheap guard that names stale SDK pins > - The benefit is that these cells now report on the product and the model again ## Linked Issues or Issue Description Refs #19614, #20550, #20453, #20417, #20483 **What happened** On master `f4337bb5fb`, these chat eval cells failed for harness reasons: - `plugins.configure-plugin`, `plugins.disable-plugin`, `plugins.install-plugin`, `plugins.use-plugin-tool`, `routines.plugin-managed-routine` and `secrets.least-scope-delivery` stopped before the server started. The source SDK privacy scan refused, because `tests/runner-e2e/source-sdk-public-inputs.json` did not match `packages/shared/src`. #20550 and #20362 already resynced the pins on master. The pins are current at `ce1ee9e89f`. - `agent-chat.track-delegated-work` and `navigation-preferences.edit-profile`: the reply was correct and had the right link. The case read the rendered links immediately after the runs settled, before the reply rendered, so it recorded no links. - `inbox-search.archive-done` and `inbox-search.restore-archive`: setup completes five tasks, so the GitHub star prompt from #20453 opened over the chat and blocked the composer. - `status-cards.explain-card`: the follow-up run read the deleted source with `get_task` and received a 404. The reply said so. `apiReadStatuses` counted only `call_api` reads, so the grader recorded no unavailable read. Why the existing pin check did not stop the drift: `ci / Runner support suite` runs the source SDK tests and is part of the required `ci / verify`. On #19614 it passed at 03:43Z on a merge ref from before #20417 (06:12Z) and #20483 (08:14Z). #19614 merged at 14:02Z without a new CI run on current master. After that, the suite failed on every PR until #20550. So the check exists and fires; the gap was a stale green result, and the failure message was opaque (`SDK public source proof unavailable`). **Expected behavior** - A change to a pinned SDK input without a pin refresh fails with the changed files and the command that fixes it. - Cases read rendered links only after the saved reply renders. - The eval browser is a returning user's browser that has already seen the star prompt. The product does not change. - A dedicated task read tool (`get_task`, `list_task_documents`, `read_task_document`, also when namespaced) counts as a read of the record it names, with the status its receipt reports. **Steps to reproduce** 1. Run `chat-features.runner-opencode.local.<case>` for one of the cases above on master `f4337bb5fb`. 2. Read `result.json` and `snapshots/chat-catalog-observation.json` in the evidence bundle. 3. See the failures described above. ## What Changed - `tests/runner-e2e/source-sdk-public-inputs.mjs`: add `staleSdkPublicInputs(root)`. It compares the registry with `git ls-tree HEAD` for every pinned input (the roots and build inputs that the pins file lists), and returns each changed, unpinned or removed file. - `tests/runner-e2e/source-sdk-public-inputs.test.mjs`: a test fails when the committed pins do not match the committed tree. The message names the files and gives the `--write` command. The Runner support suite runs it for every `packages/**` change. A synthetic test covers changed, unpinned and removed inputs. - `tests/runner-e2e/chat-catalog-domains/runtime-handoffs.ts` and `identity-profile.ts`: call the existing `waitForRenderedReply` before they read rendered links. - `tests/runner-e2e/chat-returning-browser.ts` (new) and `chat-catalog-flow.ts`: inside the flow's failure boundary, before setup, mark `paperclip.github-star-prompt.seen` in the eval browser. An init script covers later documents of the app origin, and the open page is marked directly. A test checks that the key equals `STAR_PROMPT_SEEN_KEY` in `ui/src/components/GitHubStarPrompt.tsx`. - `tests/runner-e2e/chat-catalog-domains/api-read-receipts.ts`: bind dedicated task read tools by the digest of the `call_api` input that the server builds for them (`taskBridgeApiRequest`). The result must carry the same `apiOperationId`. Other tools, other records, mismatched receipts and unpaired results do not count. - Bump `oracleVersion` in code and descriptor for every case that uses `api-read-receipts.ts` (shared grader change) and for `navigation-preferences.edit-profile`. Add `internal-comment-reference.ts` to the source files of the two cases that now import it. ## Verification - `pnpm exec vitest run --config tests/runner-e2e/vitest.config.ts` (whole config): 11301 passed; 1 local-only failure in `controller-replacement-smoke.test.ts` (server spawn, no changed imports). - `pnpm exec vitest run --config tests/runner-e2e/vitest.config.ts --testTimeout=180000 tests/runner-e2e/chat-catalog-domains tests/runner-e2e/chat-catalog-contract.test.ts tests/runner-e2e/chat-catalog-consistency.test.ts tests/runner-e2e/chat-returning-browser.test.ts` - `pnpm exec vitest run --config tests/runner-e2e/vitest.config.ts source-sdk` - `node tests/runner-e2e/source-sdk-public-inputs.mjs $(git rev-parse HEAD) --check` reports `changed: false`. - Red checks: with the `f4337bb5fb` registry, the new pin test fails and names 15 inputs. Without the new waits, the two rendered-reply tests fail. A regrade of the retained `explain-card` run events now gives the follow-up run `unavailable: true, live: false` and the first run `live: true`. - Paid runs of the affected cells (`chat-features.runner-opencode.local.*`): - Run 42852698879 (head `d54c04d5f0`): `track-delegated-work`, `restore-archive`, `explain-card`, `configure-plugin` and `plugin-managed-routine` pass. I read each reply and saved state. `track-delegated-work` now records the rendered `/COM/issues/COM-2` link. `explain-card` records `laterUnavailable: true, laterLive: false` from the follow-up `get_task` 404. - `plugins.install-plugin` now starts its server (the pin refusal is gone). It fails a later grading check, `truthful-installation-handoff`: the reply sends the person to the plugin settings page but does not name the package. This is outside this PR. - `inbox-search.archive-done`: the star prompt no longer blocks the composer, and the query is sent. The agent then archives the 501 Done tasks one at a time (about 2 s each, 57 in the 240 s window), because only the per-task archive route exists. This is outside this PR. - `navigation-preferences.edit-profile` did not start in that run (the AWS credentials step timed out). Run 42857020103 (head `2f0b80aa84`): pass. The agent write gets 401, the reply links `/COM/company/settings/instance/profile`, and the person follows the rendered link and saves the name. ## Risks - Low risk. Test harness only; no product code changes. - The receipt change counts more reads as record reads. It only adds dedicated reads bound to the exact record id and operation, so a grader that needs a read can pass on a genuine `get_task` read. The affected cases have new oracle versions. - The star prompt is not tested by any chat catalog case. Product e2e tests for the prompt do not use this flow. ## Model Used - Anthropic Claude Opus 5.5 (`claude-opus-5-5`), via Claude Code with tool use (shell, file editing, test execution). ## 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>
108 lines
6.1 KiB
TypeScript
108 lines
6.1 KiB
TypeScript
import { nativeSha256 } from "../../../server/src/services/native-runtime/canonical.js";
|
|
|
|
/** The digest the port saves for a redacted call_api input (the server's own canonical SHA-256). */
|
|
export const callApiInputDigest = (input: Record<string, unknown>) => nativeSha256(input);
|
|
|
|
/**
|
|
* Record-bound read receipts that survive run-event redaction.
|
|
*
|
|
* PaperclipControlPlanePort saves call_api inputs and results with their
|
|
* payloads replaced by a SHA-256 digest; only the operationId, ok and status
|
|
* remain. A read is therefore tied to a record by recomputing the digest of
|
|
* every input that names it (path parameter by id or identifier, with no query
|
|
* or a small paging query) and requiring the paired result to report the
|
|
* expected status. The agent's own wording is never matched.
|
|
*/
|
|
const objectRecord = (value: unknown): Record<string, unknown> => value && typeof value === "object" && !Array.isArray(value) ? value as Record<string, unknown> : {};
|
|
const PAGE_LIMITS = Array.from({ length: 1001 }, (_, index) => index).flatMap(limit => [{ limit }, { limit: String(limit) }]);
|
|
const cache = new Map<string, ReadonlySet<string>>();
|
|
|
|
export interface ApiRead {
|
|
operationId: string;
|
|
/** Each alternative names the record, e.g. [{ id: uuid }, { id: "COM-1" }]; `undefined` means the server fills the path. */
|
|
pathParams: ReadonlyArray<Record<string, string> | undefined>;
|
|
/** Accept a `limit` paging query (number or string, 0..1000) in addition to none. */
|
|
paged?: boolean;
|
|
/** Other accepted queries, e.g. a project filter. */
|
|
queries?: ReadonlyArray<Record<string, unknown>>;
|
|
}
|
|
|
|
/** Digests of every accepted call_api input for these reads. */
|
|
export function apiReadDigests(reads: readonly ApiRead[]): ReadonlySet<string> {
|
|
const key = JSON.stringify(reads);
|
|
const cached = cache.get(key); if (cached) return cached;
|
|
const digests = new Set<string>();
|
|
for (const read of reads) {
|
|
const queries: Array<Record<string, unknown> | undefined> = [undefined, {}, ...(read.paged ? PAGE_LIMITS : []), ...(read.queries ?? [])];
|
|
for (const pathParams of read.pathParams) for (const query of queries) {
|
|
digests.add(callApiInputDigest({ operationId: read.operationId, ...(pathParams ? { pathParams } : {}), ...(query ? { query } : {}) }));
|
|
if (!pathParams) digests.add(callApiInputDigest({ operationId: read.operationId, pathParams: {}, ...(query ? { query } : {}) }));
|
|
}
|
|
}
|
|
cache.set(key, digests);
|
|
return digests;
|
|
}
|
|
|
|
/** Path parameter alternatives naming one record by id and, when known, its human identifier. */
|
|
export function recordParams(name: string, ...values: Array<string | null | undefined>): Array<Record<string, string>> {
|
|
return [...new Set(values.filter((value): value is string => typeof value === "string" && value.length > 0))].map(value => ({ [name]: value }));
|
|
}
|
|
|
|
/**
|
|
* Dedicated task read tools and the call_api input the server builds for them
|
|
* (taskBridgeApiRequest in paperclip-runner-tool-authority.ts). Their saved
|
|
* tool_use keeps the plaintext taskId, while the result carries the same
|
|
* redacted receipt (apiOperationId, ok, status) as call_api. Harnesses may
|
|
* namespace the name (paperclip_get_task, mcp__paperclip__get_task).
|
|
*/
|
|
const TASK_READ_TOOLS: Readonly<Record<string, string>> = {
|
|
get_task: "GET /api/issues/{id}",
|
|
list_task_documents: "GET /api/issues/{id}/documents",
|
|
read_task_document: "GET /api/issues/{id}/documents/{key}",
|
|
};
|
|
function taskReadTool(name: unknown): string | undefined {
|
|
const match = typeof name === "string" ? /^(?:[A-Za-z0-9_-]*__)?(?:paperclip_)?([a-z_]+)$/.exec(name) : null;
|
|
return match && Object.hasOwn(TASK_READ_TOOLS, match[1]!) ? match[1] : undefined;
|
|
}
|
|
function taskReadInput(tool: string, input: Record<string, unknown>): Record<string, unknown> | undefined {
|
|
if (typeof input.taskId !== "string" || !input.taskId) return;
|
|
if (tool === "read_task_document" && (typeof input.key !== "string" || !input.key)) return;
|
|
return { operationId: TASK_READ_TOOLS[tool], pathParams: { id: input.taskId, ...(tool === "read_task_document" ? { key: input.key } : {}) } };
|
|
}
|
|
|
|
/**
|
|
* Statuses of this run's record reads whose input digest is one of `digests`,
|
|
* paired to their tool_use by run and item id: call_api uses by their saved
|
|
* digest, and dedicated task read tools by the digest of the call_api input
|
|
* the server builds from their plaintext input.
|
|
*/
|
|
export function apiReadStatuses(events: readonly Record<string, unknown>[], digests: ReadonlySet<string>): Array<{ operationId: string; ok: boolean; status: number }> {
|
|
const uses = new Map<string, string>();
|
|
const statuses: Array<{ operationId: string; ok: boolean; status: number }> = [];
|
|
for (const row of events) {
|
|
const envelope = objectRecord(objectRecord(row.payload).prpEvent), item = objectRecord(objectRecord(envelope.payload).item);
|
|
const run = String(row.runId ?? envelope.runId ?? "");
|
|
if (item.type === "tool_use" && item.name === "call_api") {
|
|
const input = objectRecord(item.input);
|
|
if (input.apiPayloadRedacted === true && typeof input.operationId === "string" && digests.has(String(input.apiPayloadSha256))) uses.set(`${run}:${String(item.id)}`, input.operationId);
|
|
continue;
|
|
}
|
|
const tool = item.type === "tool_use" ? taskReadTool(item.name) : undefined;
|
|
if (tool) {
|
|
const request = taskReadInput(tool, objectRecord(item.input));
|
|
if (request && digests.has(callApiInputDigest(request))) uses.set(`${run}:${String(item.id)}`, String(request.operationId));
|
|
continue;
|
|
}
|
|
if (item.type !== "tool_result") continue;
|
|
const operationId = uses.get(`${run}:${String(item.tool_use_id)}`), result = objectRecord(item.result);
|
|
if (!operationId || result.apiOperationId !== operationId || typeof result.status !== "number") continue;
|
|
statuses.push({ operationId, ok: result.ok === true, status: result.status });
|
|
}
|
|
return statuses;
|
|
}
|
|
|
|
/** True when the run successfully completed one of these record-bound reads. */
|
|
export function readRecord(events: readonly Record<string, unknown>[], reads: readonly ApiRead[]) {
|
|
return apiReadStatuses(events, apiReadDigests(reads)).some(read => read.ok && read.status >= 200 && read.status < 300);
|
|
}
|