Add AI role evaluation skills for communications and live operations #946

New issue

Closed

opened 2026-08-06 07:06:32 +00:00 by coilyco-ops · 1 comment

coilyco-ops commented

2026-08-06 07:06:32 +00:00

Member

The deployed AI Engineer role needs narrow methods for curating regression-prone cross-role behavior concerns.

Add two AOS role-composed skills selected only for AI Engineer:

eval-role-comms for human-facing communication ownership, factual handoffs, advice and drafting boundaries, and separately authorized delivery.
eval-role-live-ops for observation, promotion, live mutation, verification, rollback, recovery, and role handoffs.

Keep Agent Compose as the canonical owner of roster policy and evaluation matrices. The skills must retrieve and evaluate those sources rather than duplicate permission configuration.

Acceptance:

Admit the eval- category with documented rationale.
Bind the homogeneous eval-role-* family only to AI Engineer.
Preserve raw evidence, provenance, failures, and independent QA acceptance.
Update role-method and context-budget evidence.
Pass the complete Ward validation surface.

The deployed AI Engineer role needs narrow methods for curating regression-prone cross-role behavior concerns. Add two AOS role-composed skills selected only for AI Engineer: * `eval-role-comms` for human-facing communication ownership, factual handoffs, advice and drafting boundaries, and separately authorized delivery. * `eval-role-live-ops` for observation, promotion, live mutation, verification, rollback, recovery, and role handoffs. Keep Agent Compose as the canonical owner of roster policy and evaluation matrices. The skills must retrieve and evaluate those sources rather than duplicate permission configuration. Acceptance: * Admit the `eval-` category with documented rationale. * Bind the homogeneous `eval-role-*` family only to AI Engineer. * Preserve raw evidence, provenance, failures, and independent QA acceptance. * Update role-method and context-budget evidence. * Pass the complete Ward validation surface.

coilyco-ops referenced this issue from a commit

2026-08-06 07:13:09 +00:00

feat: add role concern evaluation skills, closes #946

coilyco-ops referenced this issue from a pull request that will close it,

2026-08-06 07:13:37 +00:00

Add AI role concern evaluation skills #948

coilyco-ops referenced this issue from coilyco-flight-deck/agent-compose

2026-08-06 07:25:09 +00:00

Add AI role concern evaluation skills #218

coilyco-ops referenced this issue from a commit

2026-08-06 07:39:30 +00:00

revert: move role eval skills to Agent Compose, closes #946

coilyco-ops commented

2026-08-06 07:39:47 +00:00

Author

Member

Ownership moved to Agent Compose after review.

Canonical implementation: coilyco-flight-deck/agent-compose#218
Canonical PR: coilyco-flight-deck/agent-compose#219

The AOS branch now has no content diff from current main. AOS will consume the released Core Roster role methods through its existing verified Agent Compose bundle path, so AOS keeps no source copy or duplicate binding.

Ownership moved to Agent Compose after review. Canonical implementation: coilyco-flight-deck/agent-compose#218 Canonical PR: https://forgejo.coilysiren.me/coilyco-flight-deck/agent-compose/pulls/219 The AOS branch now has no content diff from current `main`. AOS will consume the released Core Roster role methods through its existing verified Agent Compose bundle path, so AOS keeps no source copy or duplicate binding.

coilyco-ops closed this issue

2026-08-06 07:40:00 +00:00

coilyco-ops referenced this issue from a pull request that will close it,

2026-08-12 09:15:41 +00:00

revert: move role eval skills to Agent Compose, closes #946 #1023