[codex] Document terminal bench dispatch config (#4961)

## Thinking Path

> - Paperclip agents rely on skills for repeatable operating procedures
> - The Terminal-Bench loop skill needs to preserve enough dispatch
configuration to reproduce real heartbeat behavior
> - A bare benchmark command can create unassigned work with no
heartbeat-enabled agent, which is a harness setup failure rather than
product evidence
> - The Paperclip heartbeat skill also needs to keep escalation biased
toward agent-owned follow-through
> - This pull request documents dispatch runner config requirements and
strengthens the agent follow-through rule
> - The benefit is fewer misleading benchmark loops and clearer agent
operating guidance

## What Changed

- Documented `PAPERCLIP_HARBOR_RUNNER_CONFIG` / runner dispatch config
as required Terminal-Bench loop input.
- Updated the Terminal-Bench loop smoke check to require the dispatch
config mention.
- Added stronger Paperclip skill guidance to avoid asking humans for
work an agent can perform.

## Verification

- `pnpm smoke:terminal-bench-loop-skill`

## Risks

- Low risk: documentation and smoke expectation changes only. The
stricter smoke assertion is intentional so future edits do not drop the
dispatch config requirement.

> For core feature work, check [`ROADMAP.md`](ROADMAP.md) first and
discuss it in `#dev` before opening the PR. Feature PRs that overlap
with planned core work may need to be redirected — check the roadmap
first. See `CONTRIBUTING.md`.

## Model Used

- OpenAI Codex, GPT-5 coding agent, tool use and local command
execution. Exact context window was not exposed in the runtime.

## 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 run tests locally and they pass
- [x] I have added or updated tests where applicable
- [x] If this change affects the UI, I have included before/after
screenshots
- [x] I have updated relevant documentation to reflect my changes
- [x] I have considered and documented any risks above
- [x] I will address all Greptile and reviewer comments before
requesting merge

---------

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

skills/paperclip/SKILL.md

@@ -233,6 +233,10 @@ For commands, response fields, and MCP tools, read:
 - **Hiring**: use the `paperclip-create-agent` skill for new agent creation workflows (links to reusable `AGENTS.md` templates like `Coder` and `QA`).
 - **Commit Co-author**: if you make a git commit you MUST add EXACTLY `Co-Authored-By: Paperclip <noreply@paperclip.ing>` to the end of each commit message. Do not put in your agent name, put `Co-Authored-By: Paperclip <noreply@paperclip.ing>`.
 
+This is rule #1:
+
+IMPORTANT: **NEVER ASK A HUMAN TO DO WHAT AN AGENT COULD DO**. If you need to escalate, escalate. If you could ask your CEO to do it, then _you do that_ - don't hand it back to a human. Again: Never ask a human to do what an agent _could_ do. Rule number 1.
+
 ## Comment Style (Required)
 
 When posting issue comments or writing issue descriptions, use concise markdown with:
@@ -349,3 +353,5 @@ Results are ranked by relevance: title matches first, then identifier, descripti
 ## Full Reference
 
 For detailed API tables, JSON response schemas, worked examples (IC and Manager heartbeats), governance/approvals, cross-team delegation rules, error codes, issue lifecycle diagram, and the common mistakes table, read: `skills/paperclip/references/api-reference.md`
+
+Again, rule #1 is: never ask a human to do what an agent could do. Try harder. Try again. Ask another agent to help. Keep working until the goal is fully accomplished.