test(evals): bind matched baseline to built daemon evidence

Co-Authored-By: Paperclip <noreply@paperclip.ing>

packages/paperclip-runner/src/live/runnerd-codex-transport.test.ts

@@ -4154,7 +4154,9 @@ it("captures exact provider frames and correlates Rust and TypeScript interpreta
   );
   const tracePath = join(traceDirectory, "trace.ndjson");
   const bundle = createCapabilityRunnerdCodexTransport({
-    runnerBinary: defaultCapabilityRunnerdBinary(),
+    // Qualification builds a debug daemon for this exact source. Its selected
+    // binary must win over any separately staged product/runtime artifact.
+    runnerBinary: process.env.PAPERCLIP_STOCK_PREFLIGHT_RUNNERD ?? defaultCapabilityRunnerdBinary(),
     codexCommand: fakeCodex,
     codexArgs: fakeCodexArgs(traceDirectory, "--structured-activity"),
     stateDirectory: join(traceDirectory, "state"),

tests/runner-e2e/STOCK-HARNESS.md

@@ -148,6 +148,9 @@ prompt/oracle checks and the Rust additive test. It stopped before providers:
 the real daemon-frame test lacked the cold `paperclip-runnerd` binary. Setup now
 builds that daemon from the locked Rust source before TypeScript gates, retains
 `runnerd-build.txt`, and requires both setup exits in the admission receipt.
+The required daemon-frame test selects that built debug binary explicitly,
+without replacing staged product binaries. The receipt records its SHA-256;
+verification rejects a changed binary before provider admission.
 
 Dispatch the trusted workflow from `master`, with `target_branch` naming the
 same-repository candidate and an exact cell selector first. The workflow resolves

tests/runner-e2e/catalog.ts

@@ -1226,7 +1226,7 @@ export const runnerSuites: readonly RunnerSuiteFixture[] = [
     expectedMatrixSize: 24,
     definitionMetadata: {
       version: 1, instructions: "historical-default-hire-comparison", scheduling: "explicit-only",
-      comparison: { variant: "previous-instructions", candidateSha: "ac6ddefb589c18fa9c30946db40e58499e9fb0e0", restoredFrom: "e00d10d5d5594f6e2d1e8cf4e0ec81c074b0bd88", nativeCodexFix: "held-constant-14920" },
+      comparison: { variant: "previous-instructions", candidateSha: "712dc98cf52a26fafd1c894e40c081da991c4fa7", restoredFrom: "e00d10d5d5594f6e2d1e8cf4e0ec81c074b0bd88", nativeCodexFix: "held-constant-14920" },
       sourceDigest: stockHarnessSourceDigest(),
       grading: "public-default-bundle-and-delivered-prompts-plus-independent-lifecycle-oracles",
       vendorBaseEvidence: "required deterministic Codex driver/runnerd/Rust gate; task success is not vendor-base proof",

tests/runner-e2e/stock-harness-checks.mjs

@@ -89,6 +89,7 @@ export function assertPreflightReceipt(report, current) {
   if (report?.schema !== "paperclip.stock-harness-preflight.v3" || report.passed !== true ||
       report.setup?.passed !== true || report.setup?.exitCode !== 0 ||
       report.setup?.sdkExitCode !== 0 || report.setup?.runnerdExitCode !== 0 ||
+      report.setup?.runnerdSha256 !== current.runnerdSha256 ||
       report.providerCalls !== 0 || report.sourceSha !== current.sha ||
       report.sourceFingerprint !== current.fingerprint || report.sourceErrors?.length !== 0 ||
       !Array.isArray(report.gates) || report.gates.length !== expected.length ||
@@ -112,6 +113,8 @@ export function main(args = process.argv.slice(2)) {
     if (git.status !== 0 || source.sourceErrors.length) throw new Error("Cannot verify stock harness source provenance.");
     const report = assertPreflightReceipt(JSON.parse(readFileSync(verify, "utf8")), {
       sha: git.stdout.trim(), fingerprint: source.fingerprint,
+      runnerdSha256: createHash("sha256").update(readFileSync(join(root,
+        "packages/paperclip-runner/runner/target/debug", `paperclip-runnerd${process.platform === "win32" ? ".exe" : ""}`))).digest("hex"),
     });
     const output = resolve(verify, "..");
     for (const gate of stockHarnessGates) {
@@ -157,6 +160,9 @@ export function main(args = process.argv.slice(2)) {
     process.exitCode = 1;
     return;
   }
+  const runnerdBinary = join(root, "packages/paperclip-runner/runner/target/debug",
+    `paperclip-runnerd${process.platform === "win32" ? ".exe" : ""}`);
+  setup.runnerdSha256 = createHash("sha256").update(readFileSync(runnerdBinary)).digest("hex");
   const results = [];
   for (const gate of stockHarnessGates) {
     console.log(`Checking ${gate.id}: ${gate.name}`);
@@ -165,7 +171,9 @@ export function main(args = process.argv.slice(2)) {
       ...(gate.config ? ["--config", gate.config] : []),
       ...(gate.testPattern ? ["--testNamePattern", gate.testPattern] : []),
       "--reporter=default", "--reporter=json", `--outputFile.json=${file}`],
-    { cwd: resolve(root, gate.cwd), env, stdio: "inherit", timeout: 10 * 60_000 });
+    { cwd: resolve(root, gate.cwd),
+      env: gate.id === "SH-1" ? { ...env, PAPERCLIP_STOCK_PREFLIGHT_RUNNERD: runnerdBinary } : env,
+      stdio: "inherit", timeout: 10 * 60_000 });
     let report;
     try { report = JSON.parse(readFileSync(file, "utf8")); } catch { /* missing evidence fails closed below */ }
     results.push(gradeGate(gate, report, run.status));

tests/runner-e2e/stock-harness-checks.test.mjs

@@ -40,9 +40,9 @@ describe("stock harness prerequisite coverage", () => {
 });
 
 describe("stock harness prerequisite admission", () => {
-  const current = { sha: "a".repeat(40), fingerprint: "b".repeat(64) };
+  const current = { sha: "a".repeat(40), fingerprint: "b".repeat(64), runnerdSha256: "c".repeat(64) };
   const receipt = () => ({ schema: "paperclip.stock-harness-preflight.v3", passed: true,
-    setup: { passed: true, exitCode: 0, sdkExitCode: 0, runnerdExitCode: 0 },
+    setup: { passed: true, exitCode: 0, sdkExitCode: 0, runnerdExitCode: 0, runnerdSha256: current.runnerdSha256 },
     providerCalls: 0, sourceSha: current.sha, sourceFingerprint: current.fingerprint,
     sourceErrors: [], gates: [...stockHarnessGates.map(g => g.id), "SH-1-rust"].map(id => ({ id, passed: true, exitCode: 0 })) });
   it("admits the same passing source revision", () => expect(assertPreflightReceipt(receipt(), current).passed).toBe(true));
@@ -57,6 +57,7 @@ describe("stock harness prerequisite admission", () => {
     ["missing source", r => { r.sourceErrors.push("missing.ts"); }],
     ["failed cold setup", r => { r.setup.passed = false; r.setup.exitCode = 1; }],
     ["missing daemon build", r => { delete r.setup.runnerdExitCode; }],
+    ["changed daemon binary", r => { r.setup.runnerdSha256 = "d".repeat(64); }],
   ])("rejects %s before providers", (_name, mutate) => {
     const r = receipt(); mutate(r); expect(() => assertPreflightReceipt(r, current)).toThrow("exact source SHA and fingerprint");
   });

tests/runner-e2e/stock-harness-comparison.json

@@ -1,7 +1,7 @@
 {
   "schema": "paperclip.stock-harness-comparison.v1",
   "variant": "previous-instructions",
-  "candidateSha": "ac6ddefb589c18fa9c30946db40e58499e9fb0e0",
+  "candidateSha": "712dc98cf52a26fafd1c894e40c081da991c4fa7",
   "restoredInstructionsFrom": "e00d10d5d5594f6e2d1e8cf4e0ec81c074b0bd88",
   "nativeCodex14920": "held-constant; no before/after performance claim",
   "productionDifferences": [