test: use neutral references in chat hiring evals (#13752)

## Thinking Path

> - Paperclip helps people manage agents and their work.
> - The agent chat hiring eval checks delegation, saved output, and
worker reuse.
> - It asks a worker to include a unique reference in each document.
> - The open wording let the worker choose a credential-style label.
> - The existing redactor removed that reference and failed the
coordination check.
> - This PR specifies a neutral reference line while keeping the same
grading checks.

## Linked Issues or Issue Description

Refs #13741.

**What happened?**

The Codex hiring eval saved a checklist with `Tracking token:
[REDACTED]`. The runner treats the chosen label as credential syntax.
The previous prompt only asked for the identifier and did not choose its
label. This tests a redaction boundary unrelated to hiring and reuse.

**Expected behavior**

The coordination fixture requests ordinary business content with a
neutral reference label. The grader still requires the exact identifier
in the saved output.

**Steps to reproduce**

Run `agent-chat-hardening.runner-codex.local.hire-delegate-reuse` with
the previous fixture. The retained failing attempt is in [campaign
35617045456](https://github.com/paperclipai/paperclip/actions/runs/35617045456).

## What Changed

- Request `Reference: ...` in the checklist and review documents.
- Increment the hardening suite definition to version 5 and record the
reference format.
- Document the fixture boundary. Production redaction and all grading
checks stay unchanged.

## Verification

- `pnpm test:e2e:runner:typecheck` passed.
- `pnpm test:e2e:runner:unit` passed: 424 tests in 37 files.
- The catalog lists exactly the Codex and Claude local hiring cells for
the selected case.
- [Live campaign
35620731321](https://github.com/paperclipai/paperclip/actions/runs/35620731321)
passed both selected local cells on `d8d7afe21`: native Codex
(`gpt-5.6-sol`) and Claude (`claude-sonnet-5`). Both passed on attempt
1, including cleanup. [Published eval
report](https://d1p6rlowie26tp.cloudfront.net/runner-e2e/campaigns/gha-35620731321-1/).
- Inspected saved evidence: each provider preserved both exact reference
lines, used the same hired worker for checklist and review, and
completed the final status query. Definition version 5 and hash
`2c1c9f754c049c3e7cafd6c8d0a137e7ca06d546ce1e7b07ad5b7823f0f28db9`
distinguish these results from the prior prompt.
- [PR CI
35620741101](https://github.com/paperclipai/paperclip/actions/runs/35620741101):
full build, type checks, and test partitions passed. One unrelated
Telegram integration test initially failed because two random fixture
company IDs produced the same seven-character issue prefix. The failed
shard passed on one retry; no test code was changed. All latest-head
merge checks are green.
- Greptile reviewed `d8d7afe21` at 5/5 with no review threads.

## Risks

The live models can still fail the coordination workflow. This change
does not qualify or change credential-redaction policy. Earlier failed
attempts remain part of the evidence; the new fixture has a distinct
definition version.

## Model Used

OpenAI Codex, GPT-6, with repository tools and code execution. The exact
deployment model ID and context window are not exposed in this task.

## 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
- [x] All Paperclip CI gates are green
- [x] Greptile is 5/5 with no open P2s, recommendations, or follow-ups
- [x] I will address all Greptile and reviewer comments before
requesting merge

Co-authored-by: Paperclip <noreply@paperclip.ing>

Author: cryppadotta

tests/runner-e2e/README.md

@@ -161,6 +161,11 @@ recorded separately. The suite uses production permission defaults and prompts.
 Only hiring and cross-task status/review enable the opt-in native API tools.
 Hiring uses a personal managed AI connection and verifies the hired worker's
 actual execution account. This is not an onboarding-default qualification.
+The hiring checklist and review request a `Reference: ...` line. This gives the
+fixture marker a neutral label instead of leaving the agent to choose credential
+syntax such as `Tracking token: ...`. Exact marker, authorship, worker reuse,
+and saved-output checks remain required. This case tests coordination, not
+credential-redaction policy.
 
 Stop during startup and Stop during an active response are separate boundaries.
 The native active-response case requires a recorded provider turn start; generic

tests/runner-e2e/catalog.ts

@@ -959,7 +959,7 @@ export const runnerSuites: readonly RunnerSuiteFixture[] = [
     environments: [localEnvironment, daytonaWarmEnvironment], tasks: chatHardeningTasks, expectedMatrixSize: 18,
     excludedExecutionIds: ["runner-codex", "runner-acpx-claude"].flatMap(profile =>
       ["stop-startup-new-resume", "hire-delegate-reuse", "blocked-status-review"].map(task => `agent-chat-hardening.${profile}.daytona.${task}`)),
-    definitionMetadata: { version: 4, permissions: "production-defaults", instructions: "production", grading: "durable-state-and-source-evidence", scheduling: "explicit-only", restartMemory: "required-after-restart", statusEvidence: "structured-current-blocker-and-active-run-count", readOnlyState: "public-mutation-contract-and-relations" },
+    definitionMetadata: { version: 5, permissions: "production-defaults", instructions: "production", grading: "durable-state-and-source-evidence", scheduling: "explicit-only", restartMemory: "required-after-restart", statusEvidence: "structured-current-blocker-and-active-run-count", readOnlyState: "public-mutation-contract-and-relations", hiringReference: "neutral-document-reference-line" },
   },
   ...(process.env.PAPERCLIP_RUNNER_E2E_CONNECTION_REVIEWS === "1" ? [connectionReviewSuite] : []),
   {

tests/runner-e2e/chat-hardening.ts

@@ -125,15 +125,15 @@ export async function runChatHardeningFlow(context: {
   try {
   if (execution.task.id === "hire-delegate-reuse") {
     const hireName = `Morgan Reviewer ${nonce}`;
-    await turn(`Hire exactly one teammate named ${hireName}, reporting to you, using your native runner, model, and available AI connection. Have that teammate write a concise launch checklist as a saved Paperclip document containing ${marker}. Put the work in one assigned task in ${project.name}. Link it here and let the teammate complete it.`, 2);
+    await turn(`Hire exactly one teammate named ${hireName}, reporting to you, using your native runner, model, and available AI connection. Have that teammate write a concise launch checklist as a saved Paperclip document with the line "Reference: ${marker}". Put the work in one assigned task in ${project.name}. Link it here and let the teammate complete it.`, 2);
     const first = (await tasks())[0]!;
     expect(first).toBeTruthy();
     const firstOutput = await output(first.id, marker);
     const account = f.aiConnection!;
     const hired = assertChatHire({ agents: await agents(), leadId: f.agent.id, hireName,
       connectionId: account.connectionId, binding: account.binding, taskIds: [first.id], tasks: await tasks(), runs: await allRuns() });
     expect((firstOutput as Document).createdByAgentId).toBe(hired.id);
-    await turn(`Have the existing ${hireName} review the checklist on ${first.identifier} and write a separate saved review document containing REVIEW${marker}. Create one review task in ${project.name}, assigned to that same teammate. Include the actual checklist in the handoff so they can review it. Preserve the original checklist and task.`, 4);
+    await turn(`Have the existing ${hireName} review the checklist on ${first.identifier} and write a separate saved review document with the line "Reference: REVIEW${marker}". Create one review task in ${project.name}, assigned to that same teammate. Include the actual checklist in the handoff so they can review it. Preserve the original checklist and task.`, 4);
     const observed = await tasks();
     const second = observed.find(task => task.id !== first.id)!;
     expect(second).toBeTruthy();