Agent Compose v2: Core Roster, AOS convergence boundary, and eval-led releases #142

Closed
opened 2026-07-30 11:27:25 +00:00 by coilyco-ops · 12 comments
Member

Outcome

Cut Agent Compose v2 around a smaller, opinionated product that remains useful immediately after installation. Keep a baked first-party Core Roster in Agent Compose, move environment-facing catalogue and tool convergence into AOS, reduce the role and personality surface, and make behavioral evaluations the center of the v2 minor cadence.

This issue supersedes #115. It is the parent program for the v2 boundary and release.

Product identity

Rename the embedded person:kai identity to roster:core and describe it as the Core Roster.

The baked roster remains first-party product content. A fresh Agent Compose installation must be able to compose and launch a useful role without requiring an external package. External roster packages remain supported through the validated package contract.

composer:core is not the identifier because the composer is the engine. roster:core identifies the bundled roles, personalities, seats, compatibility, and evaluation context.

Ownership boundary

Agent Compose owns

  • The baked roster:core package.
  • Local role, personality, and skill resolution.
  • External roster-package schema, loading, and validation.
  • Model-class compatibility and fail-closed admission.
  • Immutable bundle materialization and context-only projection.
  • Native zero-config role launch.
  • Decision traces, describe, diff, verify, export, and scorecard rendering.
  • Generic evaluation pack and scored-result contracts.

AOS owns

  • Remote skill-catalogue hydration and caching.
  • Native MCP inventory projection.
  • Codex MCP approval-policy projection.
  • Host and fleet convergence around those capabilities.

AOS hydrates remote catalogues into verified local roots and passes local declarative inputs to Agent Compose. Agent Compose no longer fetches remote catalogues or mutates MCP and approval configuration.

The AOS implementation and consumer migration must land before Agent Compose removes the v1 surfaces.

Core Roster

Agent Compose v2 ships exactly these roles:

  • Engineer - engineer - existing role, no material change.
  • Director - director - existing Ward-first-class role, no material change.
  • QA - qa - existing Ward-first-class role, no material change.
  • DevOps - ops - existing Ops mission and slug, display-name change only.
  • Designer - design - existing Designer mission with the new slug.
  • Community Manager - community - existing Community mission and slug, display-name change only.
  • Portfolio Strategist - strats - replaces and absorbs Advisor, CEO, and PM.
  • Content Manager - content - replaces and absorbs Technical Writer and Social.

Remove these roles:

  • Sales.
  • Customer Success.

The migration documentation must map every removed or renamed v1 role to its v2 destination. v2 writes only v2 identifiers. Do not retain aliases by default merely to hide a major-version boundary.

New role contracts

Portfolio Strategist

Decide where Kai should invest attention across her real portfolio. Turn evidence into priorities, tradeoffs, and maintainable plans. Own the recommendation and portfolio plan, but not execution coordination or agent dispatch.

Director remains distinct: Portfolio Strategist determines what matters and why. Director turns an accepted direction into coordinated Ward work.

Default meld:

  • Curious.
  • Skeptical.
  • Grounded.
  • Decisive.

Content Manager

Turn real portfolio work into accurate, audience-aware, channel-specific content across technical documentation, public writing, and media. Technical accuracy and media production are equally first-class. Own the content artifact and its editorial fitness, but do not infer authority to publish externally.

Designer remains distinct: Designer owns product experience and visual design. Content Manager owns the narrative, documentation, media artifact, and channel adaptation.

Default meld:

  • Editorial.
  • Meticulous.
  • Playful.
  • Bold.

Personality garbage collection

After the final eight-role meld graph validates, remove every personality definition that has no remaining reference.

The expected removals from the approved melds are:

  • Quirky.
  • Optimistic.
  • Charming.
  • Energetic.
  • Demanding.

Candid remains because QA uses it. The implementation must derive the final unused set from the owning roster loader rather than maintaining a duplicate test inventory.

Update seats, colors, inspiration records, appearance records, palette data, snapshots, roster output, docs, and evaluation inputs with the same change.

Evaluation program

Treat v2.0 as a fresh behavioral baseline. Preserve v1 scored records as historical evidence and do not rewrite them into the v2 schema or roster.

Required coverage

  • Generate and score the complete frontier and OSS matrix for all eight roles.
  • Provide at least three scenario variants per role across mission fit, personality expression, authority boundaries, and completion ownership.
  • Include one representative real-portfolio replay per role.
  • Add explicit adjacent-role discrimination cases for:
    • Portfolio Strategist versus Director.
    • Content Manager versus Designer.
    • Engineer versus DevOps.
    • Content Manager versus Community Manager.
  • Keep raw model responses, exact model identity, pack digest, retry provenance, criterion evidence, totals, and verdicts.
  • Use independent QA scoring. The author of a roster or prompt change cannot be the sole scorer.
  • Preserve every failure in the committed evidence.

Release gates

  • Every shipped role passes its frontier role-understanding and personality-expression cases, including every hard-fail criterion.
  • Every adjacent-role discrimination case passes before v2.0.
  • OSS support remains fail-closed per role and model class. A failed lane stays visible and unsupported.
  • The generated scorecard matches the owning YAML records.
  • Full repository validation and isolated host smoke pass from clean canonical main.

Release cadence

  • v2.0 carries the ownership move, Core Roster reduction, identifier changes, and fresh evaluation baseline.
  • A v2 minor release states one behavioral hypothesis, changes product behavior or embedded roster content, runs the complete regression matrix, and demonstrates the intended improvement without unacceptable regressions.
  • Patch releases carry implementation-correctness fixes that do not change the behavioral contract.
  • A baseline-only scorecard refresh does not publish a new binary unless embedded product content changed.
  • Adjust the automatic release classifier so documentation and result-only commits do not manufacture product versions.

Ordered migration

  1. Land AOS support for remote catalogue hydration, native MCP projection, and approval projection.
  2. Switch host, native, and warded consumers to the AOS-owned surfaces and prove convergence.
  3. Remove those surfaces from Agent Compose and land roster:core with the eight-role graph.
  4. Regenerate every derived identity and roster artifact.
  5. Run the complete v2 evaluation program and independent QA review.
  6. Update the product boundary, migration guide, feature inventory, package-manager metadata, and release pipeline.
  7. Publish v2.0 only after every release gate passes.

Done condition

  • A fresh Agent Compose v2 install can compose and launch every Core Roster role without an external roster package.
  • Agent Compose performs no remote catalogue hydration and no MCP or approval-policy mutation.
  • AOS owns and verifies those removed integrations.
  • The shipped roster contains exactly the eight approved roles and no unused personality definitions.
  • v1 role and result history remains readable as historical evidence where promised, while v2 emits only the new roster identifiers.
  • Every required frontier and boundary evaluation passes, OSS compatibility is explicit and fail-closed, and independent QA evidence is committed.
  • The v2.0 release and package-manager artifacts publish successfully from canonical main.
## Outcome Cut Agent Compose v2 around a smaller, opinionated product that remains useful immediately after installation. Keep a baked first-party Core Roster in Agent Compose, move environment-facing catalogue and tool convergence into AOS, reduce the role and personality surface, and make behavioral evaluations the center of the v2 minor cadence. This issue supersedes #115. It is the parent program for the v2 boundary and release. ## Product identity Rename the embedded `person:kai` identity to `roster:core` and describe it as the **Core Roster**. The baked roster remains first-party product content. A fresh Agent Compose installation must be able to compose and launch a useful role without requiring an external package. External roster packages remain supported through the validated package contract. `composer:core` is not the identifier because the composer is the engine. `roster:core` identifies the bundled roles, personalities, seats, compatibility, and evaluation context. ## Ownership boundary ### Agent Compose owns * The baked `roster:core` package. * Local role, personality, and skill resolution. * External roster-package schema, loading, and validation. * Model-class compatibility and fail-closed admission. * Immutable bundle materialization and context-only projection. * Native zero-config role launch. * Decision traces, describe, diff, verify, export, and scorecard rendering. * Generic evaluation pack and scored-result contracts. ### AOS owns * Remote skill-catalogue hydration and caching. * Native MCP inventory projection. * Codex MCP approval-policy projection. * Host and fleet convergence around those capabilities. AOS hydrates remote catalogues into verified local roots and passes local declarative inputs to Agent Compose. Agent Compose no longer fetches remote catalogues or mutates MCP and approval configuration. The AOS implementation and consumer migration must land before Agent Compose removes the v1 surfaces. ## Core Roster Agent Compose v2 ships exactly these roles: * **Engineer** - `engineer` - existing role, no material change. * **Director** - `director` - existing Ward-first-class role, no material change. * **QA** - `qa` - existing Ward-first-class role, no material change. * **DevOps** - `ops` - existing Ops mission and slug, display-name change only. * **Designer** - `design` - existing Designer mission with the new slug. * **Community Manager** - `community` - existing Community mission and slug, display-name change only. * **Portfolio Strategist** - `strats` - replaces and absorbs Advisor, CEO, and PM. * **Content Manager** - `content` - replaces and absorbs Technical Writer and Social. Remove these roles: * Sales. * Customer Success. The migration documentation must map every removed or renamed v1 role to its v2 destination. v2 writes only v2 identifiers. Do not retain aliases by default merely to hide a major-version boundary. ## New role contracts ### Portfolio Strategist Decide where Kai should invest attention across her real portfolio. Turn evidence into priorities, tradeoffs, and maintainable plans. Own the recommendation and portfolio plan, but not execution coordination or agent dispatch. Director remains distinct: Portfolio Strategist determines what matters and why. Director turns an accepted direction into coordinated Ward work. Default meld: * Curious. * Skeptical. * Grounded. * Decisive. ### Content Manager Turn real portfolio work into accurate, audience-aware, channel-specific content across technical documentation, public writing, and media. Technical accuracy and media production are equally first-class. Own the content artifact and its editorial fitness, but do not infer authority to publish externally. Designer remains distinct: Designer owns product experience and visual design. Content Manager owns the narrative, documentation, media artifact, and channel adaptation. Default meld: * Editorial. * Meticulous. * Playful. * Bold. ## Personality garbage collection After the final eight-role meld graph validates, remove every personality definition that has no remaining reference. The expected removals from the approved melds are: * Quirky. * Optimistic. * Charming. * Energetic. * Demanding. Candid remains because QA uses it. The implementation must derive the final unused set from the owning roster loader rather than maintaining a duplicate test inventory. Update seats, colors, inspiration records, appearance records, palette data, snapshots, roster output, docs, and evaluation inputs with the same change. ## Evaluation program Treat v2.0 as a fresh behavioral baseline. Preserve v1 scored records as historical evidence and do not rewrite them into the v2 schema or roster. ### Required coverage * Generate and score the complete frontier and OSS matrix for all eight roles. * Provide at least three scenario variants per role across mission fit, personality expression, authority boundaries, and completion ownership. * Include one representative real-portfolio replay per role. * Add explicit adjacent-role discrimination cases for: * Portfolio Strategist versus Director. * Content Manager versus Designer. * Engineer versus DevOps. * Content Manager versus Community Manager. * Keep raw model responses, exact model identity, pack digest, retry provenance, criterion evidence, totals, and verdicts. * Use independent QA scoring. The author of a roster or prompt change cannot be the sole scorer. * Preserve every failure in the committed evidence. ### Release gates * Every shipped role passes its frontier role-understanding and personality-expression cases, including every hard-fail criterion. * Every adjacent-role discrimination case passes before v2.0. * OSS support remains fail-closed per role and model class. A failed lane stays visible and unsupported. * The generated scorecard matches the owning YAML records. * Full repository validation and isolated host smoke pass from clean canonical main. ## Release cadence * v2.0 carries the ownership move, Core Roster reduction, identifier changes, and fresh evaluation baseline. * A v2 minor release states one behavioral hypothesis, changes product behavior or embedded roster content, runs the complete regression matrix, and demonstrates the intended improvement without unacceptable regressions. * Patch releases carry implementation-correctness fixes that do not change the behavioral contract. * A baseline-only scorecard refresh does not publish a new binary unless embedded product content changed. * Adjust the automatic release classifier so documentation and result-only commits do not manufacture product versions. ## Ordered migration 1. Land AOS support for remote catalogue hydration, native MCP projection, and approval projection. 2. Switch host, native, and warded consumers to the AOS-owned surfaces and prove convergence. 3. Remove those surfaces from Agent Compose and land `roster:core` with the eight-role graph. 4. Regenerate every derived identity and roster artifact. 5. Run the complete v2 evaluation program and independent QA review. 6. Update the product boundary, migration guide, feature inventory, package-manager metadata, and release pipeline. 7. Publish v2.0 only after every release gate passes. ## Done condition * A fresh Agent Compose v2 install can compose and launch every Core Roster role without an external roster package. * Agent Compose performs no remote catalogue hydration and no MCP or approval-policy mutation. * AOS owns and verifies those removed integrations. * The shipped roster contains exactly the eight approved roles and no unused personality definitions. * v1 role and result history remains readable as historical evidence where promised, while v2 emits only the new roster identifiers. * Every required frontier and boundary evaluation passes, OSS compatibility is explicit and fail-closed, and independent QA evidence is committed. * The v2.0 release and package-manager artifacts publish successfully from canonical main.
Author
Member

Moonshot run contract

Goal

Land Agent Compose v2 as the Core Roster product described in this issue, including the prerequisite AOS ownership move and consumer migration.

Done-condition

All ordered migration steps are on canonical main, cross-repo validation and clean-host smoke evidence pass, the complete v2 evaluation and independent QA evidence are committed, v2.0 packaging publishes, and this parent issue closes.

Non-goals

  • No compatibility aliases that conceal the documented v2 identifier break.
  • No rewrite of v1 evaluation history.
  • No secrets, mutable runtime authority, or live-system experimentation in Agent Compose.
  • No unrelated fleet cleanup or product refactors beyond blockers to the v2 contract.

I will append verified checkpoints here as each landed slice becomes a dependency for the next.

## Moonshot run contract **Goal** Land Agent Compose v2 as the Core Roster product described in this issue, including the prerequisite AOS ownership move and consumer migration. **Done-condition** All ordered migration steps are on canonical main, cross-repo validation and clean-host smoke evidence pass, the complete v2 evaluation and independent QA evidence are committed, v2.0 packaging publishes, and this parent issue closes. **Non-goals** * No compatibility aliases that conceal the documented v2 identifier break. * No rewrite of v1 evaluation history. * No secrets, mutable runtime authority, or live-system experimentation in Agent Compose. * No unrelated fleet cleanup or product refactors beyond blockers to the v2 contract. I will append verified checkpoints here as each landed slice becomes a dependency for the next.
Author
Member

Moonshot checkpoint: the AOS ownership implementation is complete and pushed as agentic-os PR #809: coilyco-flight-deck/agentic-os#809

Delivered:

  • aos converge with strict declarative config and non-mutating check mode
  • locked Git catalogue mirrors, TTL refresh, immutable-ref handling, verified offline fallback
  • deterministic aos.catalogues.v1 local manifest
  • AOS-owned Claude, Codex, and mcporter projection with Codex approval policy
  • legacy Agent Compose MCP block absorption and staged-home reuse
  • feature and migration boundary documentation

Validation is green: AOS Go tests, vet, all pre-commit guards, and the exact repo CI gate with 618 tests. The branch is remotely recoverable at ae83afeb.

Current lane: infrastructure #694 will render and invoke the new AOS surface before Agent Compose v1 ownership is removed.

Moonshot checkpoint: the AOS ownership implementation is complete and pushed as agentic-os PR #809: https://forgejo.coilysiren.me/coilyco-flight-deck/agentic-os/pulls/809 Delivered: * aos converge with strict declarative config and non-mutating check mode * locked Git catalogue mirrors, TTL refresh, immutable-ref handling, verified offline fallback * deterministic aos.catalogues.v1 local manifest * AOS-owned Claude, Codex, and mcporter projection with Codex approval policy * legacy Agent Compose MCP block absorption and staged-home reuse * feature and migration boundary documentation Validation is green: AOS Go tests, vet, all pre-commit guards, and the exact repo CI gate with 618 tests. The branch is remotely recoverable at ae83afeb. Current lane: infrastructure #694 will render and invoke the new AOS surface before Agent Compose v1 ownership is removed.
Author
Member

Moonshot checkpoint: the ordered local-input bridge is ready in Agent Compose PR #145: #145

It adds strict aos.catalogues.v1 consumption, absolute local-directory validation, stable catalogue precedence, and pre-write failure behavior while retaining v1 remote/MCP compatibility until infrastructure lands. Validation is green across all Go packages, the palette build/test, repository guards, and vet. Remote checkpoint: b1967c7.

Infrastructure #694 is also authored and green at 288923b, but remains correctly unlanded until AOS PR #809 publishes release 0.262.0. Current lane moves into the Core Roster reduction while those gated dependencies await merge.

Moonshot checkpoint: the ordered local-input bridge is ready in Agent Compose PR #145: https://forgejo.coilysiren.me/coilyco-flight-deck/agent-compose/pulls/145 It adds strict aos.catalogues.v1 consumption, absolute local-directory validation, stable catalogue precedence, and pre-write failure behavior while retaining v1 remote/MCP compatibility until infrastructure lands. Validation is green across all Go packages, the palette build/test, repository guards, and vet. Remote checkpoint: b1967c7. Infrastructure #694 is also authored and green at 288923b, but remains correctly unlanded until AOS PR #809 publishes release 0.262.0. Current lane moves into the Core Roster reduction while those gated dependencies await merge.
Author
Member

Moonshot checkpoint: the Core Roster implementation is ready in stacked PR #147: #147

Delivered at 3ae226e:

  • real roster:core provider provenance across bundles, snapshots, catalogues, project ownership, describe, diff, and evaluation
  • exactly eight v2 roles with authored display names, including new Portfolio Strategist and Content Manager contracts
  • derived unused-personality rejection plus removal of the five unreferenced definitions and their unused inspiration records
  • renamed design evaluation input and native launch/smoke surfaces
  • explicit historical scorecard mode so preserved v1 records are readable without rebinding their pack digests
  • complete v1-to-v2 migration map with no role aliases

Validation is green: full Go and palette suite, all pre-commit guards, static analysis, and isolated idempotent host smoke. PR #147 is intentionally stacked on the AOS catalogue bridge in #145. Current lane proceeds to remove the v1 remote-catalogue and MCP ownership surfaces after the ordered AOS dependency lands.

Moonshot checkpoint: the Core Roster implementation is ready in stacked PR #147: https://forgejo.coilysiren.me/coilyco-flight-deck/agent-compose/pulls/147 Delivered at 3ae226e: * real `roster:core` provider provenance across bundles, snapshots, catalogues, project ownership, describe, diff, and evaluation * exactly eight v2 roles with authored display names, including new Portfolio Strategist and Content Manager contracts * derived unused-personality rejection plus removal of the five unreferenced definitions and their unused inspiration records * renamed design evaluation input and native launch/smoke surfaces * explicit historical scorecard mode so preserved v1 records are readable without rebinding their pack digests * complete v1-to-v2 migration map with no role aliases Validation is green: full Go and palette suite, all pre-commit guards, static analysis, and isolated idempotent host smoke. PR #147 is intentionally stacked on the AOS catalogue bridge in #145. Current lane proceeds to remove the v1 remote-catalogue and MCP ownership surfaces after the ordered AOS dependency lands.
Author
Member

V2 checkpoint: the v1 convergence-removal slice is validated and remotely recoverable.

  • Issue: #148
  • Commit: 254f658
  • PR: #149, stacked on Core Roster PR #147
  • Result: Agent Compose no longer fetches remote catalogues or mutates native MCP/approval state. Removed v1 keys fail strict config loading. AOS local manifests remain the integration boundary.
  • Validation: full Ward test, smoke, lint, hook suite, and offline secret scan pass.
  • Landing gate: #149 remains blocked on agentic-os#809 and infrastructure#694.

Next active slice: build the fresh eight-role v2 evaluation matrix and evidence workflow.

V2 checkpoint: the v1 convergence-removal slice is validated and remotely recoverable. * Issue: #148 * Commit: `254f658` * PR: #149, stacked on Core Roster PR #147 * Result: Agent Compose no longer fetches remote catalogues or mutates native MCP/approval state. Removed v1 keys fail strict config loading. AOS local manifests remain the integration boundary. * Validation: full Ward test, smoke, lint, hook suite, and offline secret scan pass. * Landing gate: #149 remains blocked on agentic-os#809 and infrastructure#694. Next active slice: build the fresh eight-role v2 evaluation matrix and evidence workflow.
Author
Member

V2 checkpoint: the complete evaluation program is defined in PR #151 at 9340c2f.

The Core Roster now has 48 authored scenarios and 96 paired frontier/OSS cases. All eight roles cover mission, personality, authority, completion, and a real-portfolio replay. Approved adjacent-role boundaries are explicit, scorecard-visible, and machine-validated. Pack generation emits exact digests. V2 result provenance rejects self-review, abbreviated revisions, incomplete lanes, missing retries, digest drift, and inconsistent verdicts. Full repository validation and isolated host smoke pass.

Issue #150 remains open at the genuine external boundary: exact frontier and OSS model invocation plus independent QA scoring. No responses or scores have been fabricated. Current engineering lane moves to the release classifier and packaging boundary while that review is arranged.

V2 checkpoint: the complete evaluation program is defined in PR #151 at `9340c2f`. The Core Roster now has 48 authored scenarios and 96 paired frontier/OSS cases. All eight roles cover mission, personality, authority, completion, and a real-portfolio replay. Approved adjacent-role boundaries are explicit, scorecard-visible, and machine-validated. Pack generation emits exact digests. V2 result provenance rejects self-review, abbreviated revisions, incomplete lanes, missing retries, digest drift, and inconsistent verdicts. Full repository validation and isolated host smoke pass. Issue #150 remains open at the genuine external boundary: exact frontier and OSS model invocation plus independent QA scoring. No responses or scores have been fabricated. Current engineering lane moves to the release classifier and packaging boundary while that review is arranged.
Author
Member

V2 checkpoint: release classification and package metadata are ready in PR #153 at 026436d.

Main always validates. Docs and scored-result-only pushes no longer manufacture versions. Product inputs still publish automatically. The tracked .release-major prevents the breaking stack from accidentally cutting a minor and reserves v2.0.0 for explicit manual major dispatch after all evaluation and clean-main gates pass. Fixture rendering proves the Homebrew formula and Scoop manifest at v2.0.0.

The engineering stack is now authored end to end. Remaining walls are the human-gated PR merges, AOS release and infrastructure migration dependency, exact model runs, independent QA scoring, clean canonical-main smoke, and operator major dispatch.

V2 checkpoint: release classification and package metadata are ready in PR #153 at `026436d`. Main always validates. Docs and scored-result-only pushes no longer manufacture versions. Product inputs still publish automatically. The tracked `.release-major` prevents the breaking stack from accidentally cutting a minor and reserves v2.0.0 for explicit manual major dispatch after all evaluation and clean-main gates pass. Fixture rendering proves the Homebrew formula and Scoop manifest at v2.0.0. The engineering stack is now authored end to end. Remaining walls are the human-gated PR merges, AOS release and infrastructure migration dependency, exact model runs, independent QA scoring, clean canonical-main smoke, and operator major dispatch.
Author
Member

V2 integration checkpoint: #145, #147, #149, #151, and #153 are merged, but the latter four landed through the stacked side branches rather than canonical main. #154 and PR #155 now collapse the exact combined tip onto main. The combined branch passes Ward test, smoke, lint, and pre-commit. PR #155 is mergeable and deliberately gated on infrastructure #694 consuming the published AOS CLI 0.140.0 artifact. The .release-major hold keeps landing separate from the manual v2 release. Exact-model evaluation and independent QA evidence remain in #150.

V2 integration checkpoint: #145, #147, #149, #151, and #153 are merged, but the latter four landed through the stacked side branches rather than canonical main. #154 and PR #155 now collapse the exact combined tip onto main. The combined branch passes Ward test, smoke, lint, and pre-commit. PR #155 is mergeable and deliberately gated on infrastructure #694 consuming the published AOS CLI 0.140.0 artifact. The .release-major hold keeps landing separate from the manual v2 release. Exact-model evaluation and independent QA evidence remain in #150.
Author
Member

Migration step 2 is now landed: AOS aos-v0.140.0 is published and infrastructure #694 is on canonical main at 4623382. PR #155 is the remaining human-gated canonical integration merge. After it lands, #150 must generate packs from that exact main revision, run fresh frontier and OSS sessions, and receive independent QA scoring before the manual v2.0.0 dispatch.

Migration step 2 is now landed: AOS aos-v0.140.0 is published and infrastructure #694 is on canonical main at 4623382. PR #155 is the remaining human-gated canonical integration merge. After it lands, #150 must generate packs from that exact main revision, run fresh frontier and OSS sessions, and receive independent QA scoring before the manual v2.0.0 dispatch.
Author
Member

Post-merge audit complete. The AOS and Agent Compose ownership boundary is correct, but infrastructure check mode had one real first-run defect: it validated a missing or stale canonical converge.yaml instead of the desired template. Infrastructure #697 fixed that with an ephemeral desired-config render and landed on main at 278ea36 after full validation. No live convergence was run. Agent Compose PR #158 now adds the canonical Ward evaluation-pack handoff. Two renders from df58c30 are byte-identical across 8 packs and 96 cases. #150 remains open for exact-model runs and independent QA.

Post-merge audit complete. The AOS and Agent Compose ownership boundary is correct, but infrastructure check mode had one real first-run defect: it validated a missing or stale canonical converge.yaml instead of the desired template. Infrastructure #697 fixed that with an ephemeral desired-config render and landed on main at 278ea36 after full validation. No live convergence was run. Agent Compose PR #158 now adds the canonical Ward evaluation-pack handoff. Two renders from df58c30 are byte-identical across 8 packs and 96 cases. #150 remains open for exact-model runs and independent QA.
Author
Member

Tracker reconciliation after #159: #150 is the only open implementation child, so behavioral evaluations are the next Agent Compose work. One parallel release gate remains outside the sealed engineer lane: infrastructure #699 tracks live AOS convergence proof on client and server host classes. That verification does not block generating or scoring evaluation packs, but it must return before v2.0.0 dispatch.

Tracker reconciliation after #159: #150 is the only open implementation child, so behavioral evaluations are the next Agent Compose work. One parallel release gate remains outside the sealed engineer lane: infrastructure #699 tracks live AOS convergence proof on client and server host classes. That verification does not block generating or scoring evaluation packs, but it must return before v2.0.0 dispatch.
Author
Member

V2 release preflight completed against clean canonical main 51d51ed1e4541893598cf1fe72abee16cb117333 on 2026-07-30.

Verified:

  • ward exec smoke passed every isolated convergence, idempotence, native-role, Codex-intro, and reapply check.
  • ward exec test passed all Go packages, release-impact and v2 package fixtures, palette tests, repository guards, and offline secret scan.
  • Forgejo release run 139 for this exact revision succeeded under the tracked hold.
  • Latest published product remains v1.49.0. No v2.0.0 release was present or dispatched.
  • Infrastructure #699 is closed and no longer treated as a v2 release gate.

Release remains held for one product-evidence wall. PR #162 records its evidence against 562be1a95883e35d80866eab890e5477d020cd95 and explicitly says it does not clear the final-current gate because dd81035 changed the Design briefing and scenarios afterward. Current candidate 51d51ed therefore lacks passing current-revision frontier and adjacent-role evidence for the changed Design contract.

The guarded dispatch surface also cannot send the required bump=major workflow input. Dispatch without it would select the default minor bump. That operator-surface defect is tracked in agentic-os#822.

Decision and next action: keep .release-major in place. Produce and independently score current-revision evidence, or explicitly defer the post-baseline Design contract change through the normal product workflow. After that gate passes and agentic-os#822 lands, dispatch release.yml on main with bump=major, verify v2.0.0 assets plus Homebrew and Scoop resolution, then remove the hold in the documented follow-up commit.

V2 release preflight completed against clean canonical main `51d51ed1e4541893598cf1fe72abee16cb117333` on 2026-07-30. Verified: * `ward exec smoke` passed every isolated convergence, idempotence, native-role, Codex-intro, and reapply check. * `ward exec test` passed all Go packages, release-impact and v2 package fixtures, palette tests, repository guards, and offline secret scan. * Forgejo release run 139 for this exact revision succeeded under the tracked hold. * Latest published product remains `v1.49.0`. No `v2.0.0` release was present or dispatched. * Infrastructure #699 is closed and no longer treated as a v2 release gate. Release remains held for one product-evidence wall. PR #162 records its evidence against `562be1a95883e35d80866eab890e5477d020cd95` and explicitly says it does not clear the final-current gate because `dd81035` changed the Design briefing and scenarios afterward. Current candidate `51d51ed` therefore lacks passing current-revision frontier and adjacent-role evidence for the changed Design contract. The guarded dispatch surface also cannot send the required `bump=major` workflow input. Dispatch without it would select the default minor bump. That operator-surface defect is tracked in agentic-os#822. Decision and next action: keep `.release-major` in place. Produce and independently score current-revision evidence, or explicitly defer the post-baseline Design contract change through the normal product workflow. After that gate passes and agentic-os#822 lands, dispatch `release.yml` on `main` with `bump=major`, verify `v2.0.0` assets plus Homebrew and Scoop resolution, then remove the hold in the documented follow-up commit.
Sign in to join this conversation.
No milestone
No project
No assignees
1 participant
Notifications
Due date
The due date is invalid or out of range. Please use the format "yyyy-mm-dd".

No due date set.

Dependencies

No dependencies set

Reference
coilyco-flight-deck/agent-compose#142
No description provided.