mirror of
https://github.com/paperclipai/paperclip.git
synced 2026-11-15 01:52:55 +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>
101 lines
7.6 KiB
TypeScript
101 lines
7.6 KiB
TypeScript
import { expect, it } from "vitest";
|
|
import { stripTypeScriptTypes } from "node:module";
|
|
import { createHash } from "node:crypto";
|
|
import { startOpenCodeMcpBridge } from "../drivers/opencode/mcp-bridge.js";
|
|
import { readFile } from "node:fs/promises";
|
|
import { createOpenCodeFinishDiagnosticForwarder } from "./opencode-proxy-finish-diagnostics.js";
|
|
import { createOpenCodeProxyOutput } from "./opencode-proxy-output.js";
|
|
const offered = { schema: "paperclip.opencode.finish-input-diagnostic/v1", stage: "offered", schemaSha256: "a".repeat(64) };
|
|
const rejected = { ...offered, stage: "input_rejected", callIdSha256: "b".repeat(64), validationCategory: "continuation", rootShape: "object", continuationShape: "string" };
|
|
it("retains closed diagnostics through normal warning output before an empty-response failure", async () => {
|
|
const frames: unknown[] = [], output = createOpenCodeProxyOutput(frame => frames.push(frame));
|
|
const forward = createOpenCodeFinishDiagnosticForwarder(output.send);
|
|
forward.diagnostic(JSON.stringify(offered));
|
|
expect(frames).toEqual([]);
|
|
expect(forward.bind("actual-session")).toBe(true);
|
|
expect(frames).toEqual([]);
|
|
forward.diagnostic(JSON.stringify(rejected));
|
|
output.send({ method: "turn/completed", params: { threadId: "actual-session", turn: { status: "failed", error: { code: "opencode_empty_response" } } } });
|
|
expect(frames).toEqual([
|
|
{ method: "warning", params: { threadId: "actual-session", message: JSON.stringify(offered) } },
|
|
{ method: "warning", params: { threadId: "actual-session", message: JSON.stringify(rejected) } },
|
|
expect.objectContaining({ method: "turn/completed" }),
|
|
]);
|
|
const source = await readFile(new URL("./opencode-app-server-proxy.ts", import.meta.url), "utf8");
|
|
expect(source).toContain("finishDiagnostics.diagnostic(message)");
|
|
expect(source).toContain("finishDiagnostics.bind(session.ids().driverSessionId)");
|
|
expect(source).toContain('process.stderr.write(`[opencode] ${message}\\n`)');
|
|
});
|
|
it("rejects unbound rejections, foreign binding/schema, unknown fields and arbitrary text", () => {
|
|
const frames: unknown[] = [], forward = createOpenCodeFinishDiagnosticForwarder(frame => frames.push(frame));
|
|
forward.diagnostic(JSON.stringify(rejected));
|
|
forward.diagnostic("arbitrary private text");
|
|
forward.diagnostic(JSON.stringify({ ...offered, threadId: "foreign" }));
|
|
expect(frames).toEqual([]);
|
|
forward.diagnostic(JSON.stringify(offered));
|
|
expect(forward.bind("actual-session")).toBe(true);
|
|
expect(forward.bind("foreign-session")).toBe(false);
|
|
for (const invalid of [{ ...rejected, schemaSha256: "c".repeat(64) }, { ...rejected, input: "private" }, { ...rejected, rootShape: {} }, { ...rejected, rootShape: ["object"] }, { ...rejected, continuationShape: "invalid" }]) forward.diagnostic(JSON.stringify(invalid));
|
|
expect(frames).toHaveLength(0);
|
|
for (let i = 0; i < 12; i++) forward.diagnostic(JSON.stringify(rejected));
|
|
forward.diagnostic(JSON.stringify(offered));
|
|
expect(frames).toHaveLength(9);
|
|
expect(JSON.stringify(frames)).not.toMatch(/private|foreign/);
|
|
});
|
|
|
|
it("forwards from the actual proxy open callback instead of losing normal semantic failure diagnostics to stderr", async () => {
|
|
const source=await readFile(new URL("./opencode-app-server-proxy.ts",import.meta.url),"utf8");
|
|
const body=stripTypeScriptTypes(source.slice(source.indexOf("async function open("),source.indexOf("function threadResponse(")));
|
|
const frames:unknown[]=[],stderr:string[]=[],output=createOpenCodeProxyOutput(frame=>frames.push(frame));
|
|
let diagnostic:(message:string)=>void=()=>{};
|
|
class Driver {
|
|
constructor(options:{onDiagnostic:(message:string)=>void}) { diagnostic=options.onDiagnostic; }
|
|
async openSession() { diagnostic(JSON.stringify(offered));return {ids:()=>({driverSessionId:"actual-session",providerSessionId:"actual-session"})}; }
|
|
}
|
|
const invoke=new Function("OpenCodeServerDriver","createOpenCodeFinishDiagnosticForwarder","send","process",`
|
|
let session=null,driver,eventPump,cwd,activeModel,activeCompletionContract;
|
|
const openCodeProxyCompletionContract=()=>undefined;
|
|
const activeRunId="run",launchBinding={},resolve=x=>x,text=(x,f="")=>typeof x==="string"?x:f,record=x=>x;
|
|
const runtimeDirectory=()=>"/synthetic",withoutAmbientOpenCodeCommand=()=>({}),parseOpenCodeProxyPermissionMode=()=>"allow",openCodeProxyTaskEnvelope=()=>({});
|
|
const openCodeProxyCompletionFeedback=()=>()=>{},requestController=()=>{},pumpEvents=async()=>{},failProxy=()=>{},threadResponse=id=>({thread:{id}});
|
|
${body}
|
|
return open({model:"provider/model",cwd:"/synthetic"},false);
|
|
`);
|
|
await invoke(Driver,createOpenCodeFinishDiagnosticForwarder,output.send,{env:{},cwd:()=>"/synthetic",stderr:{write:(message:string)=>stderr.push(message)}});
|
|
expect(frames).toEqual([]);
|
|
diagnostic(JSON.stringify(rejected));
|
|
diagnostic("unrelated private diagnostic");
|
|
output.send({method:"turn/completed",params:{threadId:"actual-session",turn:{status:"failed",error:{code:"opencode_empty_response"}}}});
|
|
expect(frames).toHaveLength(3);
|
|
expect(frames.slice(0,2)).toEqual([offered,rejected].map(value=>({method:"warning",params:{threadId:"actual-session",message:JSON.stringify(value)}})));
|
|
expect(stderr).toHaveLength(3);
|
|
expect(JSON.stringify(frames)).not.toContain("unrelated private diagnostic");
|
|
});
|
|
|
|
it("keeps healthy open and success silent", () => {
|
|
const frames:unknown[]=[],output=createOpenCodeProxyOutput(frame=>frames.push(frame));
|
|
const forward=createOpenCodeFinishDiagnosticForwarder(output.send);
|
|
forward.diagnostic(JSON.stringify(offered));forward.bind("healthy-session");
|
|
output.send({method:"turn/completed",params:{threadId:"healthy-session",turn:{status:"completed"}}});
|
|
expect(frames).toEqual([{method:"turn/completed",params:{threadId:"healthy-session",turn:{status:"completed"}}}]);
|
|
});
|
|
|
|
it("retains the actual MCP producer's offered schema and rejected shape records", async () => {
|
|
const frames:Array<{params:{message:string}}>=[];
|
|
const forward=createOpenCodeFinishDiagnosticForwarder(frame=>frames.push(frame));
|
|
const bridge=await startOpenCodeMcpBridge({handler:async()=>({accepted:true}),onDiagnostic:forward.diagnostic});
|
|
const rpc=async(id:number,method:string,params?:unknown)=>(await fetch(bridge.url,{method:"POST",headers:{Authorization:`Bearer ${bridge.secret}`,"Content-Type":"application/json"},body:JSON.stringify({jsonrpc:"2.0",id,method,params})})).json() as Promise<{result:{tools:Array<{name:string;inputSchema:unknown}>;isError?:boolean}}>;
|
|
try {
|
|
expect(frames).toEqual([]);forward.bind("actual-session");expect(frames).toEqual([]);
|
|
const tools=await rpc(1,"tools/list"),schema=tools.result.tools.find((tool:{name:string})=>tool.name==="paperclip_finish")!.inputSchema;
|
|
const report={reportedWorkDisposition:"done",summary:"Synthetic complete response",completionClaim:{contractRevision:"1",objectiveSatisfied:true,criteria:[],remainingWork:[]},evidence:[],verification:[]};
|
|
const result=await rpc(2,"tools/call",{name:"paperclip_finish",arguments:{...report,continuation:"SYNTHETIC_PRIVATE_INPUT"}});
|
|
expect(result.result.isError).toBe(true);
|
|
expect(frames).toHaveLength(2);
|
|
const [actualOffered,actualRejected]=frames.map(frame=>JSON.parse(frame.params.message));
|
|
expect(actualOffered).toEqual({...offered,schemaSha256:createHash("sha256").update(JSON.stringify(schema)).digest("hex")});
|
|
expect(actualRejected).toEqual({...actualOffered,stage:"input_rejected",callIdSha256:createHash("sha256").update("2").digest("hex"),validationCategory:"continuation",rootShape:"object",continuationShape:"string"});
|
|
expect(JSON.stringify(frames)).not.toContain("SYNTHETIC_PRIVATE_INPUT");
|
|
} finally {await bridge.close();}
|
|
});
|