Consume renamed and split aos ward-specs bundle files

coilyco-ops commented

2026-07-10 10:11:56 +00:00

Member

Problem

The aos ward-specs bundle is being reshaped in coilyco-flight-deck/agentic-os#431, and ward is the runtime consumer of that bundle through WARD_CONFIG_REF.

Known upstream changes requested in aos:

.ward/ward-kdl.fleet.kdl -> .ward/agents.kdl, with root node agents.
.ward/ward-kdl.roles.kdl -> .ward/roles.kdl.
Role guardfile bindings move into roles.kdl as repeated singular guardfile "..." nodes.
.ward/ward-kdl.defaults.kdl -> .ward/defaults.kdl, with root node defaults.
repo-authority moves into .ward/repos.kdl, with root node repos.
Guardfile bundle members drop the ward-kdl. filename prefix and start with guardfile., for example guardfile.forgejo.write.kdl.

The cli-guard parser side is tracked separately in coilyco-flight-deck/cli-guard#219 for agents and repeated singular guardfile syntax. Ward still needs its bundle discovery/loading layer checked and updated for the renamed/split files.

Scope

Audit and update ward's WARD_CONFIG_REF bundle consumer for the new aos layout:

Locate any hard-coded member names like ward-kdl.fleet.kdl, ward-kdl.roles.kdl, ward-kdl.defaults.kdl, or ward-kdl.*.guardfile.kdl.
Update the loader to read agents.kdl, roles.kdl, defaults.kdl, repos.kdl, and guardfile.*.kdl names.
Update any parser call sites that expect roles to be embedded in the agent roster file.
Update smart defaults parsing for defaults { ... } if ward currently expects smart-defaults { ... }.
Update repo authority parsing for repos { ... } in its own file if ward currently expects repo-authority { ... } inside the defaults file.
Coordinate with cli-guard#219 for the fleetconfig syntax change. If ward vendors or pins cli-guard, bump it after cli-guard lands or make a compatibility path explicit.
Add or update tests with a fixture matching the new aos bundle layout.

Acceptance criteria

Ward can load a WARD_CONFIG_REF bundle using the new aos file names and root nodes.
Ward no longer requires roles to be embedded in the agent roster file.
Ward reads defaults and repo authority from separate defaults.kdl and repos.kdl files.
Tests cover the new bundle layout and fail on the old hard-coded assumptions.
The ward test suite passes.

ward.workflow: pull-requests-and-merge

## Problem The aos ward-specs bundle is being reshaped in `coilyco-flight-deck/agentic-os#431`, and ward is the runtime consumer of that bundle through `WARD_CONFIG_REF`. Known upstream changes requested in aos: * `.ward/ward-kdl.fleet.kdl` -> `.ward/agents.kdl`, with root node `agents`. * `.ward/ward-kdl.roles.kdl` -> `.ward/roles.kdl`. * Role guardfile bindings move into `roles.kdl` as repeated singular `guardfile "..."` nodes. * `.ward/ward-kdl.defaults.kdl` -> `.ward/defaults.kdl`, with root node `defaults`. * `repo-authority` moves into `.ward/repos.kdl`, with root node `repos`. * Guardfile bundle members drop the `ward-kdl.` filename prefix and start with `guardfile.`, for example `guardfile.forgejo.write.kdl`. The cli-guard parser side is tracked separately in `coilyco-flight-deck/cli-guard#219` for `agents` and repeated singular `guardfile` syntax. Ward still needs its bundle discovery/loading layer checked and updated for the renamed/split files. ## Scope Audit and update ward's `WARD_CONFIG_REF` bundle consumer for the new aos layout: * Locate any hard-coded member names like `ward-kdl.fleet.kdl`, `ward-kdl.roles.kdl`, `ward-kdl.defaults.kdl`, or `ward-kdl.*.guardfile.kdl`. * Update the loader to read `agents.kdl`, `roles.kdl`, `defaults.kdl`, `repos.kdl`, and `guardfile.*.kdl` names. * Update any parser call sites that expect roles to be embedded in the agent roster file. * Update smart defaults parsing for `defaults { ... }` if ward currently expects `smart-defaults { ... }`. * Update repo authority parsing for `repos { ... }` in its own file if ward currently expects `repo-authority { ... }` inside the defaults file. * Coordinate with cli-guard#219 for the fleetconfig syntax change. If ward vendors or pins cli-guard, bump it after cli-guard lands or make a compatibility path explicit. * Add or update tests with a fixture matching the new aos bundle layout. ## Acceptance criteria * Ward can load a `WARD_CONFIG_REF` bundle using the new aos file names and root nodes. * Ward no longer requires roles to be embedded in the agent roster file. * Ward reads defaults and repo authority from separate `defaults.kdl` and `repos.kdl` files. * Tests cover the new bundle layout and fail on the old hard-coded assumptions. * The ward test suite passes. ward.workflow: pull-requests-and-merge

coilyco-ops commented

2026-07-10 10:12:05 +00:00

Author

Member

WARD-DISPATCH: deferred ⏸

deferred details

This forwarded dispatch was deferred after the issue was already reserved.

Attempted harness: codex
Attempted run: ward agent engineer coilyco-flight-deck/ward#1009 --harness codex --ward-version v0.579.0
Container: engineer-codex-ward-1009
Container created: no running engineer was observed.
Host log: /Users/kai/.ward/agent-logs/dispatch/20260710T101202Z-director-codex-rh98-coilyco-flight-deck-ward-1009.log
Capacity: ward agent engineer --harness codex: global engineer limit is reached: 12 running (limit 12); wait for a run to finish or run ward agent reap for stale engineers

Retry: the issue stays queued and the director will try again when a slot opens.

— Codex, via ward agent

WARD-DISPATCH: deferred ⏸ <details><summary>deferred details</summary> This forwarded dispatch was deferred after the issue was already reserved. Attempted harness: `codex` Attempted run: `ward agent engineer coilyco-flight-deck/ward#1009 --harness codex --ward-version v0.579.0` Container: `engineer-codex-ward-1009` Container created: no running engineer was observed. Host log: `/Users/kai/.ward/agent-logs/dispatch/20260710T101202Z-director-codex-rh98-coilyco-flight-deck-ward-1009.log` Capacity: `ward agent engineer --harness codex: global engineer limit is reached: 12 running (limit 12); wait for a run to finish or run `ward agent reap` for stale engineers` Retry: the issue stays queued and the director will try again when a slot opens. </details>  — Codex, via `ward agent`

coilyco-ops commented

2026-07-10 10:13:22 +00:00

Author

Member

Additional constraint from Kai, 2026-07-10:

Kai intentionally deleted a bunch of KDL specgen from the ward repo because that code did not belong there. Keep tests passing without bringing the specgen back en masse. Prefer updating ward's current bundle-consumer code and tests around the reduced surface, not restoring the removed generator stack as the easy path.

Additional constraint from Kai, 2026-07-10: Kai intentionally deleted a bunch of KDL specgen from the ward repo because that code did not belong there. Keep tests passing without bringing the specgen back en masse. Prefer updating ward's current bundle-consumer code and tests around the reduced surface, not restoring the removed generator stack as the easy path.

coilyco-ops added the

headless

P2

labels

2026-07-10 10:17:42 +00:00

coilyco-ops commented

2026-07-10 10:18:57 +00:00

Author

Member

WARD-DISPATCH: failed ❌

failure details

This forwarded dispatch failed after the issue was already reserved.

Attempted harness: codex
Attempted run: ward agent engineer coilyco-flight-deck/ward#1009 --harness codex --details Implement ward#1009 only after checking the current aos#431 and cli-guard#219 contracts. Update ward's WARD_CONFIG_REF bundle consumer for renamed/split files and repeated singular guardfile entries, but do not bring back the deleted KDL specgen stack.
Container: engineer-codex-ward-1009
Container created: no running engineer was observed.
Host log: /Users/kai/.ward/agent-logs/dispatch/20260710T101854Z-director-codex-zv99-coilyco-flight-deck-ward-1009.log
Failure: ward container: download Go bootstrap binary https://forgejo.coilysiren.me/coilyco-flight-deck/ward/releases/download/v0.580.0/ward-linux-arm64: unexpected status 404 Not Found: Not Found

Retry: choose another harness if the first one is down, or rerun with --force if the reservation is stale.

— Codex, via ward agent

WARD-DISPATCH: failed ❌ <details><summary>failure details</summary> This forwarded dispatch failed after the issue was already reserved. Attempted harness: `codex` Attempted run: `ward agent engineer coilyco-flight-deck/ward#1009 --harness codex --details Implement ward#1009 only after checking the current aos#431 and cli-guard#219 contracts. Update ward's WARD_CONFIG_REF bundle consumer for renamed/split files and repeated singular guardfile entries, but do not bring back the deleted KDL specgen stack.` Container: `engineer-codex-ward-1009` Container created: no running engineer was observed. Host log: `/Users/kai/.ward/agent-logs/dispatch/20260710T101854Z-director-codex-zv99-coilyco-flight-deck-ward-1009.log` Failure: `ward container: download Go bootstrap binary https://forgejo.coilysiren.me/coilyco-flight-deck/ward/releases/download/v0.580.0/ward-linux-arm64: unexpected status 404 Not Found: Not Found` Retry: choose another harness if the first one is down, or rerun with `--force` if the reservation is stale. </details>  — Codex, via `ward agent`

coilyco-ops commented

2026-07-10 10:20:57 +00:00

Author

Member

WARD-RESERVATION: held 🔒

reservation details

Holder: container engineer-codex-ward-1009 on host kais-macbook-pro-2.local.

Reserved by ward agent --harness codex (reserved 2026-07-10T10:20:56Z). Concurrent ward agent runs are blocked until it finishes or the reservation goes stale (1h TTL). --force overrides.

Do not comment on or edit this issue to steer the run while it is reserved. The engineer seeded the body once at launch and never re-reads it, so a comment or edit reaches only human readers, never the running engineer. A correction goes to a new issue, dispatched fresh. That is the only channel that reaches a run in flight. Where the forge supports it, ward locks this conversation to make that a road-block rather than a convention (ward#494).

run seed context — what this run is carrying (ward#609)

Resolved: coilyco-flight-deck/ward#1009 · branch issue-1009 · harness codex · workflow pull-requests-and-merge
Run: engineer-codex-ward-1009 · ward v0.579.0 · dispatched 2026-07-10T10:20:56Z
Comment thread: 3 included in the pre-flight read, 0 stripped (ward's own automated comments).
- included: @coilyco-ops (2026-07-10T10:12:05Z), @coilyco-ops (2026-07-10T10:13:22Z), @coilyco-ops (2026-07-10T10:18:57Z)

Static container doctrine and seed boilerplate are identical every run and omitted here (they ride ward v0.579.0).

— Codex, via ward agent

WARD-RESERVATION: held 🔒 <details><summary>reservation details</summary> Holder: container `engineer-codex-ward-1009` on host `kais-macbook-pro-2.local`. Reserved by `ward agent --harness codex` (reserved 2026-07-10T10:20:56Z). Concurrent `ward agent` runs are blocked until it finishes or the reservation goes stale (1h TTL). `--force` overrides. **Do not comment on or edit this issue to steer the run while it is reserved.** The engineer seeded the body once at launch and never re-reads it, so a comment or edit reaches only human readers, never the running engineer. A correction goes to a **new issue, dispatched fresh**. That is the only channel that reaches a run in flight. Where the forge supports it, ward locks this conversation to make that a road-block rather than a convention (ward#494). <details><summary>run seed context — what this run is carrying (ward#609)</summary> - **Resolved:** `coilyco-flight-deck/ward#1009` · branch `issue-1009` · harness `codex` · workflow `pull-requests-and-merge` - **Run:** `engineer-codex-ward-1009` · ward `v0.579.0` · dispatched `2026-07-10T10:20:56Z` - **Comment thread:** 3 included in the pre-flight read, 0 stripped (ward's own automated comments). - included: @coilyco-ops (2026-07-10T10:12:05Z), @coilyco-ops (2026-07-10T10:13:22Z), @coilyco-ops (2026-07-10T10:18:57Z) Static container doctrine and seed boilerplate are identical every run and omitted here (they ride ward v0.579.0). </details> </details>  — Codex, via `ward agent`

coilyco-ops commented

2026-07-10 10:37:47 +00:00

Author

Member

Additional live evidence from the director surface on 2026-07-10 UTC:

A fresh scratch clone of current ward main at 142b3c29622a208d30b577dc77709e12c93c8946 fails ward exec test ./cmd/ward without any feature-branch changes. The failures prove this is not only future aos layout work, it is current main CI breakage after the ward-side specgen/source bundle deletion.

Representative failing checks:

TestCapabilityGuardfilesExist still expects ../../.ward/ward-kdl/ward-kdl.aws.guardfile.kdl and ward-kdl.tailscale.guardfile.kdl.
TestDefaultsAssetsMirrorWardKDL, TestFleetAssetsMirrorWardKDL, TestTopologyAssetsMirrorWardKDL, and TestOpsAssetsMatchWardKDL still read source files under ../../.ward/ward-kdl/ward-kdl.* that no longer exist in ward.
TestExecAssetsMirrorWardKDL reports embedded cmd/ward/execassets/ward-kdl.*.guardfile.kdl copies no longer match any current source guardfile and then says no exec-dialect source guardfiles were discovered.
TestRunDoctorWithValidBundle fails because the ops bundle expects ward-kdl.forgejo.guardfile.kdl.
TestRunSetupWithFixtureRef, TestSmartDefaultsBaked, and related config-source checks are also coupled to the old source layout and external bundle ref.

Current ward .ward contains only .ward/ward-kdl/README.md, .ward/ward-kdl/ward-kdl.roles.kdl, and .ward/ward.yaml, so a fix must update/delete the old mirror/source-drift assumptions. Do not restore the removed personal specgen/source tree just to satisfy these tests. Make the ward test suite green against the reduced ward repo plus the external WARD_CONFIG_REF bundle contract.

Additional live evidence from the director surface on 2026-07-10 UTC: A fresh scratch clone of current ward `main` at `142b3c29622a208d30b577dc77709e12c93c8946` fails `ward exec test ./cmd/ward` without any feature-branch changes. The failures prove this is not only future aos layout work, it is current main CI breakage after the ward-side specgen/source bundle deletion. Representative failing checks: * `TestCapabilityGuardfilesExist` still expects `../../.ward/ward-kdl/ward-kdl.aws.guardfile.kdl` and `ward-kdl.tailscale.guardfile.kdl`. * `TestDefaultsAssetsMirrorWardKDL`, `TestFleetAssetsMirrorWardKDL`, `TestTopologyAssetsMirrorWardKDL`, and `TestOpsAssetsMatchWardKDL` still read source files under `../../.ward/ward-kdl/ward-kdl.*` that no longer exist in ward. * `TestExecAssetsMirrorWardKDL` reports embedded `cmd/ward/execassets/ward-kdl.*.guardfile.kdl` copies no longer match any current source guardfile and then says no exec-dialect source guardfiles were discovered. * `TestRunDoctorWithValidBundle` fails because the ops bundle expects `ward-kdl.forgejo.guardfile.kdl`. * `TestRunSetupWithFixtureRef`, `TestSmartDefaultsBaked`, and related config-source checks are also coupled to the old source layout and external bundle ref. Current ward `.ward` contains only `.ward/ward-kdl/README.md`, `.ward/ward-kdl/ward-kdl.roles.kdl`, and `.ward/ward.yaml`, so a fix must update/delete the old mirror/source-drift assumptions. Do **not** restore the removed personal specgen/source tree just to satisfy these tests. Make the ward test suite green against the reduced ward repo plus the external `WARD_CONFIG_REF` bundle contract.

coilyco-ops commented

2026-07-10 10:37:58 +00:00

Author

Member

WARD-RESERVATION: held 🔒

reservation details

Holder: container engineer-codex-ward-1009 on host kais-macbook-pro-2.local.

Reserved by ward agent --harness codex (reserved 2026-07-10T10:37:58Z). Concurrent ward agent runs are blocked until it finishes or the reservation goes stale (1h TTL). --force overrides.

Do not comment on or edit this issue to steer the run while it is reserved. The engineer seeded the body once at launch and never re-reads it, so a comment or edit reaches only human readers, never the running engineer. A correction goes to a new issue, dispatched fresh. That is the only channel that reaches a run in flight. Where the forge supports it, ward locks this conversation to make that a road-block rather than a convention (ward#494).

run seed context — what this run is carrying (ward#609)

Resolved: coilyco-flight-deck/ward#1009 · branch issue-1009 · harness codex · workflow pull-requests-and-merge
Run: engineer-codex-ward-1009 · ward v0.580.0 · dispatched 2026-07-10T10:37:58Z
Comment thread: 4 included in the pre-flight read, 1 stripped (ward's own automated comments).
- included: @coilyco-ops (2026-07-10T10:12:05Z), @coilyco-ops (2026-07-10T10:13:22Z), @coilyco-ops (2026-07-10T10:18:57Z), @coilyco-ops (2026-07-10T10:37:47Z)
- stripped: @coilyco-ops (2026-07-10T10:20:57Z)

Static container doctrine and seed boilerplate are identical every run and omitted here (they ride ward v0.580.0).

— Codex, via ward agent

WARD-RESERVATION: held 🔒 <details><summary>reservation details</summary> Holder: container `engineer-codex-ward-1009` on host `kais-macbook-pro-2.local`. Reserved by `ward agent --harness codex` (reserved 2026-07-10T10:37:58Z). Concurrent `ward agent` runs are blocked until it finishes or the reservation goes stale (1h TTL). `--force` overrides. **Do not comment on or edit this issue to steer the run while it is reserved.** The engineer seeded the body once at launch and never re-reads it, so a comment or edit reaches only human readers, never the running engineer. A correction goes to a **new issue, dispatched fresh**. That is the only channel that reaches a run in flight. Where the forge supports it, ward locks this conversation to make that a road-block rather than a convention (ward#494). <details><summary>run seed context — what this run is carrying (ward#609)</summary> - **Resolved:** `coilyco-flight-deck/ward#1009` · branch `issue-1009` · harness `codex` · workflow `pull-requests-and-merge` - **Run:** `engineer-codex-ward-1009` · ward `v0.580.0` · dispatched `2026-07-10T10:37:58Z` - **Comment thread:** 4 included in the pre-flight read, 1 stripped (ward's own automated comments). - included: @coilyco-ops (2026-07-10T10:12:05Z), @coilyco-ops (2026-07-10T10:13:22Z), @coilyco-ops (2026-07-10T10:18:57Z), @coilyco-ops (2026-07-10T10:37:47Z) - stripped: @coilyco-ops (2026-07-10T10:20:57Z) Static container doctrine and seed boilerplate are identical every run and omitted here (they ride ward v0.580.0). </details> </details>  — Codex, via `ward agent`

coilyco-ops referenced this issue from coilyco-flight-deck/agentic-os

2026-07-10 10:49:59 +00:00

Expose guarded Forgejo Actions task rerun command #434

coilyco-ops referenced this issue

2026-07-10 10:52:53 +00:00

Replace unmergeable ward#1012 config-bundle PR with a landing fix #1014

coilyco-ops referenced this issue

2026-07-10 10:56:51 +00:00

Keep director surfaces from running out of writable test space #1008

coilyco-ops referenced this issue

2026-07-10 10:56:51 +00:00

Keep director surfaces from running out of writable test space #1024

coilyco-ops referenced this issue

2026-07-10 10:57:16 +00:00

Stop director dispatch from exceeding the global engineer limit #1016

coilyco-ops referenced this issue

2026-07-10 10:57:16 +00:00

Stop director dispatch from exceeding the global engineer limit #1025

coilyco-ops referenced this issue

2026-07-10 10:58:15 +00:00

Replace closed-unmerged ward#1019 dirty-only Goose residual PR #1021

coilyco-ops referenced this issue

2026-07-10 10:58:15 +00:00

fix(goose): dirty-only residual landing #1026

coilyco-ops referenced this issue

2026-07-10 10:59:24 +00:00

Replace unmergeable ward#1017 issue-scoped director PR #1020

coilyco-ops referenced this issue

2026-07-10 10:59:24 +00:00

fix: issue-scoped director fail closed #1027

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

2026-07-10 11:13:10 +00:00

Consume renamed and split aos ward-specs bundle files #1028

coilyco-ops closed this issue

2026-07-10 11:15:10 +00:00

coilyco-ops commented

2026-07-10 11:15:28 +00:00

Author

Member

WARD-OUTCOME: merge-ready

details

workflow: pull-requests-and-merge; review summary: skipped (review gate skipped by ~/.ward/config.yaml default)
felt: mechanical. main work was bundle-path rewiring and fixture cleanup.
confidence: high
surprises: the blank-ref smart-default cache needed to ignore target metadata; the PR CI queued a bit before going green.
follow-ups: none

WARD-OUTCOME: merge-ready <details><summary>details</summary> workflow: pull-requests-and-merge; review summary: skipped (review gate skipped by ~/.ward/config.yaml default) felt: mechanical. main work was bundle-path rewiring and fixture cleanup. confidence: high surprises: the blank-ref smart-default cache needed to ignore target metadata; the PR CI queued a bit before going green. follow-ups: none </details>

~~coilyco-ops referenced this issue 2026-07-10 18:29:14 +00:00~~

Open-PR backpressure gate: cap net-new engineers above N open PR branches (default 6), broker + pre-flight #1073

coilyco-ops referenced this issue

2026-07-10 18:30:16 +00:00

Implement issue-scoped director behavior, not parse-only acceptance #988

coilyco-ops referenced this issue

2026-07-10 18:30:16 +00:00

Replace closed-unmerged dirty-only Goose residual landing PR #1006

coilyco-ops referenced this issue

2026-07-10 18:30:40 +00:00

[epic] Stop the warded-workflow full stop: landability gate + convergence circuit-breaker + merge queue #1069