Coworker: qualify the private pilot and prepare the v1.3 adoption documentation

[jeffhuber]

Part of #868. Step C7.

Depends on: #869, #870, #871, #872, #873, #874

Target: Coworker integration for v1.3.

Problem

A usable connection needs evidence that it improves a real supervised work loop, preserves account isolation and works from both default orchestrator hosts before it is recommended in a release.

Scope

• Run a bounded two-work-item private pilot using the operator-configured account and existing tracker authority. Keep its identity, repository, ticket bodies and source context out of this issue and all public artifacts. Use frozen reference cases if implementation is already underway or completed; coordinate through the pilot owner rather than competing for its checkout.
• Compare baseline and enriched context for intended behavior, missing constraints, owner questions, useful regression tests, time to reviewed PR and observed usage/spend. Record causal limits of a two-task pilot and retain unknown values.
• Verify both Claude and Codex host flows, account mismatch, revocation, interruption/resume and required-context review failure. Finish generic install/upgrade/troubleshooting docs, local/offline regression checks and the normal package/release qualification handoff.

Acceptance criteria

• The pilot produces cited useful evidence and independently reviewed PR results, or an honest disposition explaining why it did not add value.
• Public scorecard uses only synthetic/anonymized data and allowlisted metadata, with no operator email or private context.
• Default no-provider cold install and configured-context operation pass; every implementation PR has independent current-head Code Mower audit evidence and a passing required gate. Graphify remains a separate candidate, not a release prerequisite.

Implementation anchors

• docs/planning-work-orders.md
• docs/sessions.md
• release/adoption qualification
• private pilot evidence and public sanitized scorecard

Verification

Full relevant CONTRIBUTING checks and source/package rehearsal at the release boundary; bounded live canaries and sanitized manual validation notes, with no repeated paid work to fill a quota.

Code Mower delivery

Use the parent epic's architecture and privacy contract. One assigned builder and one writer per branch; use an isolated checkout and normal session lease. Open a small issue-linked PR, run relevant offline checks, record provenance, and obtain an independent current-head Code Mower audit with the author excluded. Fix blocking findings and pass normal CI plus the required code-mower/gate before merge. Context providers do not acquire tracker-write or merge authority.

Keep account identities, private repository/ticket details, raw context and authentication output out of public issues, fixtures and PRs. Use generic examples. Never substitute another account to unblock a live probe.

Delivery status: complete. Release PR #887 passed independent Code Mower review and merged; v1.3.0 is published and installed. Final evidence follows.

Final release evidence

• v1.3.0 release and PyPI 1.3.0 are published. Release PR release: v1.3.0 optional Coworker context #887 merged as 97af705639c6fbe5b846e900b15728a451463f44; its tree matches the independently reviewed head.
• Independent current-head Claude audit: PASS, no findings. Required Code Mower gate and Python 3.12–3.14/package CI passed. Full local regression: 2,795 tests passed; privacy, Ruff, compile, no-provider smoke, cold clone and package rehearsal passed.
• Published-package campaign campaign-v1.3.0 is complete. Both required maintained adapters executed: Claude PASS; Codex PASS_WITH_WARNINGS (one installation warning, zero required owner actions). Both ended on 1.3.0.
• Published-package upgrade rehearsal from 1.2.2 to 1.3.0 completed: PASS_WITH_WARNINGS (eight doctor warnings, zero required owner actions). Package installation, Board, lane status and overhead steps passed. Warning details are not retained by the allowlisted result schema.
• Local installation now uses code-mower[coworker]==1.3.0. The installed CLI's explicit online connection diagnostic reports ready/verified_online. Code Mower's own Board was refreshed and reports healthy on 1.3.0.
• Sanitized pilot scorecard: two bounded frozen-reference cases; identical authorized evidence delivered across Claude/Codex orchestrator, builder and reviewer roles. One related-control consistency question was corroborated; the other case added no actionable constraint. This was not a new live private-PR gate trial or a measured productivity gain. Private identities, source evidence and owner follow-up remain local.
• Optional cloud upload remains unconfigured and off. No raw private pilot data was published. Graphify Context: evaluate official Graphify as a local repository provider for v1.4.1 #876 remains a later candidate.

All C1–C7 deliverables are complete; no release blocker or required owner action remains.