# Cross-Harness Relay: Grok Bot, Claude Code, Codex and Hermes on one job

> A bot template and teaching guide for running more than one agent harness on the same work. Six seat cards (front desk, planner, builder, reviewer, tester, night shift), a RELAY.md seat map, a relay packet for every handoff, four loop shapes, routine-or-judgment routing with a fallback when limits run tight, a preflight check, the front desk and night shift split with a proof footer, shared memory rules, eight relay drills, a rules block and 14 prompts.

- URL: https://paperchasewebb.ai/bots/marketplace/items/cross-harness-relay-grok-bot-claude-code-codex-hermes-seats-handoffs/
- Type: Bot template
- Category: Vibe coding
- By: PaperChaseWebb
- Works with: Claude Code, Codex, Grok Bot, Hermes (named in the listing); Plain .md
- Price: $39 USD
- Difficulty: Intermediate
- Setup time: 1 hr 30 min
- Time to result: A filled RELAY.md, seat cards in place and the eight drills passing on a test repo in one afternoon
- Format: Markdown (.md) file, cross-harness-relay-grok-bot-claude-code-codex-hermes-seats-handoffs.md
- Version: 1.0, updated 2026-10-27
- License: Personal and commercial use on your own projects and your team's, including setups you build for clients. Do not resell or republish this pack.

## What you end up with

Every agent knows its seat and reach, no agent approves its own work, each handoff carries the context the next one lacks, and unattended jobs stop and ask instead of guessing.

## What you get

- 14 sections in one .md file
- Seats, not models: a seat table with reach and exit
- A RELAY.md seat map template
- Six seat cards to paste into instructions
- The relay packet and when you read it
- Four loop shapes with a picker table
- Routine-or-judgment routing, a limit fallback and per-seat model rules
- A preflight check with three error classes
- Front desk and night shift split with a proof footer
- Shared memory rules, a failure table and eight relay drills
- A rules block and 14 prompts

## Who it is for

Builders already using two or more agents who want them working one job without lost context, self-approval or surprise actions.

## Tools needed

- Grok Bot: front desk: takes requests, relays approvals, reports results
- Claude Code or Codex: planner, builder or reviewer seats, read from RELAY.md
- Hermes Agent: night shift: unattended jobs in a narrow profile
- You need: Two or more agent harnesses you already use (for example Grok Bot, Claude Code, Codex or Hermes)
- You need: A repo with tests, or a test repo for the drills
- You need: An owner who reads the plan and merges

Cost to run: The pack is the only purchase. Uses the agent accounts you already have.

## How it works

1. **Pick a loop shape.** Use prompt 2 and the table in section 6. Start with two seats unless you can name the problem a third one fixes. Expected: One shape for the project.
2. **Fill RELAY.md.** Use prompt 1. Write reach and 'may not' for each seat before deciding which harness fills it. Expected: A seat map every harness reads first.
3. **Paste the seat cards and rules block.** Into CLAUDE.md, AGENTS.md, your Grok Bot instructions and each Hermes profile. Expected: Each agent knows its seat, reach and exit.
4. **Add the preflight and proof footer.** Use prompts 11 and 12 for any seat that uses tools or runs unattended. Expected: Jobs stop before they start when a tool cannot act.
5. **Run the eight drills.** On a test repo. Each must pass before real work goes through the relay. Expected: A pass log you trust.

## Sample prompt: Draft the seat map (Claude Code, Codex, Grok Bot)

```
Read this repo and my notes [paste]. Draft RELAY.md using this template [paste]. Leave 'Filled by' blank.
For each seat, propose the narrowest reach that still lets it do the job, and list what it may not do.
```

Decide who fills each seat last.

## Sample prompt: Cross-model review (Claude Code, Codex)

```
You did not write this diff. Compare it to PLAN.md [paste].
List findings as high, medium or low with file, line and a one-line fix. Say plainly if nothing is high.
```

Use a different model from the builder.

## Free preview of cross-harness-relay-grok-bot-claude-code-codex-hermes-seats-handoffs.md

````markdown
# Cross-Harness Relay: Grok Bot, Claude Code, Codex and Hermes on one job

A bot template and teaching guide by PaperChaseWebb for running more than one agent harness on the same piece of work. Seat cards for a front desk, planner, builder, reviewer, tester and night shift, a RELAY.md seat map, a relay packet every handoff carries, routing rules, a preflight check, shared memory rules, eight relay drills, a rules block and 14 prompts. Plain .md.

This is a method, not a hosted product and not an endorsement of any vendor. You bring the accounts you already use. The pack tells each agent which seat it sits in, what it may touch, what it must hand on, and where you read before the work moves.

## Contents

1. Why run more than one harness, and when not to
2. Seats, not models
3. The seat map: RELAY.md
4. Six seat cards
5. The relay packet
6. Four loop shapes
7. Routing rules: routine or judgment
8. Preflight: prove the next seat can act
9. Front desk and night shift
10. Shared memory between harnesses
11. Failure modes and fixes
12. Eight relay drills
13. Rules block and 14 prompts
14. Sources

## 1. Why run more than one harness, and when not to

Builders who use Grok Bot, Claude Code, Codex and Hermes side by side keep landing on the same few reasons:

- **A second reader.** A model reviewing its own diff tends to agree with itself. A different model, or a different seat with a different brief, finds the step most likely to fail.
- **Different jobs, different homes.** Conversation and approvals belong with whatever you talk to all day. Long unattended jobs belong somewhere built to run them. Routine code belongs wherever it costs least to produce and is easiest to review.
- **Limits.** Quotas and usage caps run out at bad times. A written fallback route keeps work moving without the agent deciding on its own where to send it.

When not to: if one agent with a good rules file and a test suite already gets the job done, a relay adds handoffs, and every handoff is a place to lose context. Start with two seats (builder and reviewer). Add a seat only when you can name the problem it fixes.

## 2. Seats, not models

Write the team as seats. A seat is a job with a brief, a reach (what it may read and change) and an exit (what it must hand on). Any harness can fill any seat. You decide who fills which seat from your own trial runs, not from a leaderboard.

| Seat | Job | Typical reach | Exit |
|---|---|---|---|
| Front desk | Takes requests, asks for approvals, reports results | Chat, calendar, read access to the repo and logs | A relay packet to the next seat, or a report to you |
| Planner | Turns a ticket into a plan with files, tests and risks | Read the repo, write PLAN.md | A plan you approve |
| Builder | Implements one approved plan | A branch, the test runner | A diff with tests green |
| Reviewer | Reviews the diff against the plan | Read the branch, write REVIEW.md | Findings by severity |
| Tester | Uses the running app like a person | A test or staging URL, a test account | Bugs written as a spec |
| Night shift | Runs long or scheduled jobs unattended | A narrow profile with only the tools the job needs | A result with a proof footer |

Two rules hold across every seat: no seat approves its own work, and no seat widens its own reach.
````

## Worth knowing

- Product names, plan limits, subagent features and APIs change often. Check each vendor's current docs and terms before scripting around them.
- A relay adds handoffs. If one agent with tests already does the job, keep it simple.
- Unattended jobs need tight limits. Keep anything irreversible behind a person.

## Questions

**Do I need all four harnesses?** No. The pack is written as seats, and any harness can fill any seat. Two seats (builder and reviewer) is the recommended start.

**How is this different from Grok Bot Templates or Governed Agent Teams?** Those cover one bot's personas and routines, and authority inside one team. This pack is about moving one job between different harnesses: the seat map, the relay packet, routing, preflight and the front desk and night shift split.

**Does it tell me which model is best?** No. You fill seats from your own trial runs. The drills tell you whether a seat is working.

## How to buy

$39 one time at https://paperchasewebb.ai/bots/marketplace/items/cross-harness-relay-grok-bot-claude-code-codex-hermes-seats-handoffs/. Stripe checkout, then an instant download page; the download link is also in your emailed receipt.

Seller: PaperChaseWebb Inc., mgmt@paperchasewebb.com