@open-agent-toolkit/cli 0.3.2 → 0.3.3

package/assets/bundle-metadata.json

@@ -1,4 +1,4 @@
 {
   "schemaVersion": 1,
-  "oatVersion": "0.3.2"
+  "oatVersion": "0.3.3"
 }

package/assets/config/dispatch-matrix-recommendation.json

@@ -1,5 +1,5 @@
 {
-  "version": "2026-09-23.2",
+  "version": "2026-09-24.1",
   "providers": {
     "codex": {
       "economy": {
@@ -63,8 +63,13 @@
           },
           {
             "harness": "codex",
-            "model": "gpt-6-sol",
-            "effort": "max"
+            "model": "gpt-6-astra",
+            "effort": "high"
+          },
+          {
+            "harness": "codex",
+            "model": "gpt-6-astra",
+            "effort": "xhigh"
           }
         ]
       }

package/assets/docs/cli-utilities/configuration.md

@@ -475,8 +475,9 @@ even when the reusable ladder is user-owned.
 The example above is illustrative rather than a copy of the bundled ladder; its
 tiers are trimmed for readability.
 
-The bundled recommendation covers 10 Codex model/effort combinations: GPT-6
-Luna at `low` through `max`, and GPT-6 Sol at the same efforts.
+The bundled recommendation covers 11 Codex model/effort combinations: GPT-6
+Luna at `low` through `max`, GPT-6 Sol at `low` through `high` plus `xhigh`,
+and GPT-6 Astra at `high` and `xhigh`.
 Claude covers `haiku`, plus explicit Sonnet, Opus, and Fable
 model/effort pairs. Effort-pinned Claude cells require a recognized versioned
 model ID (`fable-5-1`, `fable-5`, `opus-5-5`, `sonnet-5`,

package/assets/docs/contributing/updating-model-guidance.md

@@ -34,8 +34,12 @@ probe records can still describe past behavior, clearly labeled as historical.
 
 Check the provider's official model documentation and the local runtime before
 editing an ID or effort range. The [OpenAI model catalog](https://developers.openai.com/api/docs/models)
-lists GPT-6 Sol and Luna; compare it with the installed Codex runtime model
-cache for the exact names and local effort set. The
+lists GPT-6 Astra, Sol, and Luna. The
+[Astra model page](https://developers.openai.com/api/docs/models/gpt-6-astra)
+names `gpt-6-astra` and documents `low`, `medium`, `high`, `xhigh`, and `max`;
+compare it with the installed Codex runtime model cache for the exact names and
+local effort set. Catalog presence establishes support, not relative task
+quality. The
 [Anthropic Opus 5.5 documentation](https://www.anthropic.com/claude-opus-5-5)
 names `claude-opus-5-5`; confirm a Claude Code invocation reports that exact
 assistant model before admitting it to the Claude catalog. Claude subagent
@@ -74,6 +78,9 @@ for the distinction between catalog, configured mapping, and runtime identity.
 2. Update the bundled recommendation's ordered candidate cells and version.
    Check the _last_ candidate in each tier: it determines that tier's pinned
    self-review target. Keep the ladder within its provider's supported catalog.
+   A recommendation may reflect an explicit operator preference before a
+   workload comparison exists; label that boundary instead of presenting the
+   preference as measured superiority or accepted portfolio policy.
 3. Update the provider selection references and bump the canonical skill
    version once for the PR. Search the repository for the old and new IDs,
    including schemas, fixtures, examples, and generated projections. Replace

package/assets/docs/provider-sync/providers.md

@@ -58,7 +58,7 @@ description: 'Provider-specific path mappings for Claude, Cursor, Copilot, Gemin
       - `.codex/agents/<role>.toml`
       - `.codex/config.toml` (`[features] multi_agent = true`, `[agents.<role>]` upserts)
     - Codex role files include OAT managed provenance headers and are regenerated by `oat sync --scope project|user|all`
-    - Project sync writes and maintains the version-controlled supported catalogue: Luna and Terra at `low`, `medium`, `high`, and `xhigh`, plus Sol at those efforts and `max`, for both `oat-phase-implementer` and `oat-reviewer` (26 pinned variants). Sync writes files; it does not create a Git commit.
+    - Project sync writes and maintains the version-controlled supported catalogue: GPT-6 Luna, Sol, and Astra at `low`, `medium`, `high`, `xhigh`, and `max`; GPT-5.6 Luna and Terra through `xhigh`; and GPT-5.6 Sol through `max`. Both `oat-phase-implementer` and `oat-reviewer` receive each supported pair (56 pinned variants). Sync writes files; it does not create a Git commit.
     - Custom targets follow configuration provenance: user-config targets write only under `~/.codex`; shared, repo-local, and active-project targets write under the project's tracked `.codex` view.
     - User sync loads only the bundled canonical `oat-phase-implementer` and `oat-reviewer` definitions needed to expand user-config targets. It does not enable general user-agent mirroring, and it refuses stale user-role cleanup if either managed definition is unavailable.
     - Generated variants carry `supported-catalogue`, `user-config`, or `project-config` ownership. Cleanup removes stale entries only for the owner being reconciled and preserves other scopes plus unrelated Codex entries.
@@ -66,7 +66,7 @@ description: 'Provider-specific path mappings for Claude, Cursor, Copilot, Gemin
     - A single role can be materialized directly with `oat providers codex materialize <agent-name> --model <model-id> --effort <reasoning-effort>`; `--agent-path` selects a specific canonical markdown agent, `--role-name` overrides the generated role name, and `--scope user` writes a user-config-owned role.
     - Aggregate Codex config drift metadata (`aggregateConfigHash`) is emitted in sync/status codex extension output and intentionally not stored as a separate manifest row
     - Sync-time materialization is best effort; managed workflow correctness uses exact registered roles or a fresh child pinned to the resolved model plus reasoning effort with canonical role instructions. After a successful provider-visible file change, OAT conservatively advises starting a new Codex session so it has an opportunity to load the role. This repository policy is not a Codex hot-reload contract, an application-process restart requirement, or proof that the new session exposed the role.
-    - Codex `max` is a first-class dispatch effort. It is present only for the Sol family in the committed supported catalogue, for both implementer and reviewer roles.
+    - Codex `max` is a first-class dispatch effort. It is supported for GPT-6 Luna, Sol, and Astra, plus GPT-5.6 Sol, for both implementer and reviewer roles.
     - Codex multi-agent dispatch uses config-defined roles (`[agents.<name>]`) and `agent_type`
     - Codex subagent workflows require `[features] multi_agent = true` in active Codex config layers
     - Default managed Codex execution requires root (depth 0) → phase implementer (depth 1). `agents.max_depth >= 2` enables optional nested phase-agent work; sync and direct materialization still merge that capability floor without lowering a higher target value.

package/assets/docs/workflows/projects/dispatch-ceiling.md

@@ -80,8 +80,11 @@ To pick up a new version, compare your
 `workflow.dispatchCeiling.recommendationVersion` against the bundled version,
 then either edit the affected cells by hand or clear them and re-adopt.
 
-Version `2026-09-23.2` is the current bundled recommendation. It prefers
-GPT-6 Luna and Sol in Codex and Opus 5.5 low/medium/high in Claude High.
+Version `2026-09-24.1` is the current bundled recommendation. It prefers
+GPT-6 Luna and Sol in Codex, then Sol xhigh, Astra high, and Astra xhigh in
+Codex Frontier. Astra's task advantage has not been measured locally; this is
+an explicit user-directed preference, not acceptance by the separately
+maintained model-selection policy. Opus 5.5 low/medium/high remains in Claude High.
 The Claude low option suits bounded intelligent recon in Balanced or High;
 medium remains the implementation starting point under High. GPT-5.6 targets remain
 supported for explicit Codex and Cursor configurations. Cursor's preferred
@@ -234,8 +237,10 @@ oat_dispatch_policy:
 The bundled ladder is a curated subset of the supported targets:
 
 - **Codex:** GPT-6 Luna at `low` through `max` across Economy and Balanced;
-  GPT-6 Sol at `low` through `max` across High and Frontier. GPT-5.6
-  variants remain available outside this preferred ladder.
+  GPT-6 Sol at `low` through `high` in High and `xhigh` in Frontier; then
+  GPT-6 Astra at `high` and `xhigh` in Frontier. Astra and Sol remain supported
+  at `low`, `medium`, `high`, `xhigh`, and `max`, while GPT-5.6 variants remain
+  available outside this preferred ladder.
 - **Claude:** `haiku` and Sonnet 5 at Economy/Balanced, Opus 5.5 at High and
   Frontier, and Fable 5.1 at Frontier.
 - **Cursor:** the preferred ladder uses approved Composer 2.5, GPT-5.6, Grok

package/assets/public-package-versions.json

@@ -1,6 +1,6 @@
 {
-  "cli": "0.3.2",
-  "docs-config": "0.3.2",
-  "docs-theme": "0.3.2",
-  "docs-transforms": "0.3.2"
+  "cli": "0.3.3",
+  "docs-config": "0.3.3",
+  "docs-theme": "0.3.3",
+  "docs-transforms": "0.3.3"
 }

package/assets/skills/oat-project-plan-writing/SKILL.md

@@ -5,7 +5,7 @@ disable-model-invocation: true
 user-invocable: false
 allowed-tools: Read, Write, Glob, Grep
 metadata:
-  version: 1.2.30
+  version: 1.2.31
 ---
 
 # Plan Writing Contract
@@ -121,7 +121,7 @@ asking to write anything:
 
 | Provider        | Economy                                                   | Balanced                                     | High                                                              | Frontier                                                               |
 | --------------- | --------------------------------------------------------- | -------------------------------------------- | ----------------------------------------------------------------- | ---------------------------------------------------------------------- |
-| Codex           | gpt-6-luna/low, gpt-6-luna/medium, gpt-6-luna/high        | gpt-6-luna/xhigh, gpt-6-luna/max             | gpt-6-sol/low, gpt-6-sol/medium, gpt-6-sol/high                   | gpt-6-sol/xhigh, gpt-6-sol/max                                         |
+| Codex           | gpt-6-luna/low, gpt-6-luna/medium, gpt-6-luna/high        | gpt-6-luna/xhigh, gpt-6-luna/max             | gpt-6-sol/low, gpt-6-sol/medium, gpt-6-sol/high                   | gpt-6-sol/xhigh, gpt-6-astra/high, gpt-6-astra/xhigh                   |
 | Claude          | haiku, claude-sonnet-5/medium                             | claude-opus-5-5/low                          | claude-opus-5-5/low, claude-opus-5-5/medium, claude-opus-5-5/high | claude-opus-5-5/xhigh, claude-opus-5-5/max, claude-fable-5-1/high      |
 | Cursor (opaque) | `composer-2.5`, `gpt-5.6-luna-high`, `gpt-5.6-luna-xhigh` | `cursor-grok-4.5-high`, `gpt-5.6-terra-high` | `gpt-5.6-sol-medium`, `gpt-5.6-sol-high`                          | `gpt-5.6-sol-xhigh`, `gpt-5.6-sol-max`, `claude-fable-5-thinking-high` |
 

package/assets/skills/subagent-orchestration/SKILL.md

@@ -4,7 +4,7 @@ description: Use when delegating work to subagents or choosing a model for a tas
 compatibility: Self-contained; no OAT installation required.
 user-invocable: true
 metadata:
-  version: 1.0.8
+  version: 1.0.9
 ---
 
 # Subagent Orchestration

package/assets/skills/subagent-orchestration/references/evidence-and-refresh.md

@@ -1,8 +1,8 @@
 # Evidence and Refresh Policy
 
 ```yaml
-guidance_version: 2026-09-23
-last_verified: 2026-09-23
+guidance_version: 2026-09-24
+last_verified: 2026-09-24
 review_after: 2026-10-08
 stale_after: 2026-10-23
 ```
@@ -78,6 +78,11 @@ publish a provider-independent effort conversion.
 
 ## Current Evidence Summary
 
+- GPT-6 Astra is selectable in the current Codex catalogue at low through max.
+  The bundled Frontier preference uses Astra high and xhigh after Sol xhigh by
+  explicit user direction. No local task evaluation establishes an Astra
+  advantage, and this preference is not acceptance by the separately
+  maintained model-selection policy.
 - GPT-6 Sol and Luna are selectable in the current Codex catalogue and are
   the default new-work routes in `provider-codex.md`. GPT-5.6 Sol/Luna/Terra
   remain supported for explicit configurations. The July independent coding

package/assets/skills/subagent-orchestration/references/provider-codex.md

@@ -1,6 +1,6 @@
 ---
-guidance_version: 2026-09-22
-last_verified: 2026-09-22
+guidance_version: 2026-09-24
+last_verified: 2026-09-24
 review_after: 2026-12-22
 ---
 
@@ -13,11 +13,23 @@ model examples below. Launch mechanics for OAT dispatch live in
 
 ## Current Families
 
-Use the verified GPT-6 Sol/Luna model IDs for new Codex work:
+Use verified GPT-6 Sol/Luna model IDs for routine new Codex work. GPT-6 Astra
+is a supported Frontier candidate:
 
+- `gpt-6-astra`: Frontier candidate for especially demanding work; the bundled
+  recommendation uses high and xhigh, while the supported catalogue also
+  permits low, medium, and max;
 - `gpt-6-sol`: high-capability implementation and reasoning route;
 - `gpt-6-luna`: cost-sensitive, bounded work.
 
+Public engineering benchmarks show task-dependent Astra gains over Sol, with
+cost and runtime tradeoffs; they do not establish a universal replacement. The
+September frontier-release comparison remains review-pending in the separate
+model-selection vault, and no matched OAT workload evaluation has qualified a
+routine default change. Astra's Frontier inclusion is a user-directed
+preference. Keep the Sol task-class defaults below while that evidence is
+reconciled and local task fit is evaluated.
+
 The GPT-5.6 Sol/Luna/Terra targets remain supported for explicit configurations
 and established workloads. Re-evaluate economics and qualitative routing on
 representative work before carrying over GPT-5.6-specific comparisons.
@@ -47,7 +59,10 @@ model/effort pairs are in the local Codex catalogue.
 | `hard-reasoning`         | `gpt-6-sol`, high                          | Sol xhigh                                                             | Narrow the problem before raising effort.                                     |
 | `consequential`          | `gpt-6-sol`, high, plus independent review | Sol xhigh for a reasoning-depth bottleneck; max only after evaluation | The root retains consequential authorization.                                 |
 
-The bundled frontier ladder and the supported Sol effort set end at max.
+The bundled Frontier ladder ends with Astra xhigh after Sol xhigh and Astra
+high. The supported Astra and Sol effort sets both end at max, so Sol max
+remains available for explicit configuration even though it is no longer in
+the bundled Frontier recommendation.
 GPT-5.6 `reasoning.mode: "pro"` is a
 separate quality/latency/cost control on that older model, not a GPT-6 effort
 label or model slug.

package/dist/providers/codex/codec/shared.d.ts

@@ -8,6 +8,9 @@ export declare const SUPPORTED_CODEX_ROLE_TARGETS: readonly ({
 } | {
     model: string;
     effort: string;
+} | {
+    model: string;
+    effort: string;
 } | {
     model: string;
     effort: "high" | "low" | "medium" | "xhigh";

package/dist/providers/codex/codec/shared.d.ts.map

@@ -1 +1 @@
-{"version":3,"file":"shared.d.ts","sourceRoot":"","sources":["../../../../src/providers/codex/codec/shared.ts"],"names":[],"mappings":"AAIA,eAAO,MAAM,uBAAuB,wBAAwB,CAAC;AAC7D,eAAO,MAAM,4BAA4B,iBAAiB,CAAC;AAC3D,eAAO,MAAM,6BAA6B,kBAAkB,CAAC;AAE7D,MAAM,MAAM,cAAc,GACtB,qBAAqB,GACrB,aAAa,GACb,gBAAgB,CAAC;AAoBrB,eAAO,MAAM,4BAA4B;;;;;;;;;;;;;;;IAqB/B,CAAC;AAEX,wBAAgB,qBAAqB,CAAC,KAAK,EAAE,MAAM,GAAG,MAAM,CAE3D;AAED,wBAAgB,sBAAsB,CAAC,KAAK,EAAE,MAAM,GAAG,MAAM,CAO5D;AAQD,wBAAgB,oCAAoC,CAAC,OAAO,EAAE;IAC5D,SAAS,EAAE,MAAM,CAAC;IAClB,KAAK,EAAE,MAAM,CAAC;IACd,MAAM,EAAE,MAAM,CAAC;CAChB,GAAG,MAAM,CAiBT;AAgDD,wBAAgB,yBAAyB,CACvC,OAAO,EAAE,MAAM,EACf,QAAQ,CAAC,EAAE,MAAM,GAChB,OAAO,CAOT;AAED,wBAAgB,yBAAyB,CACvC,QAAQ,EAAE,MAAM,EAChB,QAAQ,EAAE,MAAM,GACf,MAAM,CAER;AAED,wBAAgB,4BAA4B,CAC1C,OAAO,EAAE,MAAM,GACd,cAAc,GAAG,IAAI,CAEvB;AAED,wBAAgB,4BAA4B,CAC1C,OAAO,EAAE,MAAM,EACf,KAAK,EAAE,cAAc,GACpB,MAAM,CAkBR;AAED,wBAAgB,aAAa,CAAC,MAAM,EAAE,MAAM,CAAC,MAAM,EAAE,OAAO,CAAC,GAAG,MAAM,CAErE"}
+{"version":3,"file":"shared.d.ts","sourceRoot":"","sources":["../../../../src/providers/codex/codec/shared.ts"],"names":[],"mappings":"AAIA,eAAO,MAAM,uBAAuB,wBAAwB,CAAC;AAC7D,eAAO,MAAM,4BAA4B,iBAAiB,CAAC;AAC3D,eAAO,MAAM,6BAA6B,kBAAkB,CAAC;AAE7D,MAAM,MAAM,cAAc,GACtB,qBAAqB,GACrB,aAAa,GACb,gBAAgB,CAAC;AAoBrB,eAAO,MAAM,4BAA4B;;;;;;;;;;;;;;;;;;IAyB/B,CAAC;AAEX,wBAAgB,qBAAqB,CAAC,KAAK,EAAE,MAAM,GAAG,MAAM,CAE3D;AAED,wBAAgB,sBAAsB,CAAC,KAAK,EAAE,MAAM,GAAG,MAAM,CAO5D;AAQD,wBAAgB,oCAAoC,CAAC,OAAO,EAAE;IAC5D,SAAS,EAAE,MAAM,CAAC;IAClB,KAAK,EAAE,MAAM,CAAC;IACd,MAAM,EAAE,MAAM,CAAC;CAChB,GAAG,MAAM,CAiBT;AAgDD,wBAAgB,yBAAyB,CACvC,OAAO,EAAE,MAAM,EACf,QAAQ,CAAC,EAAE,MAAM,GAChB,OAAO,CAOT;AAED,wBAAgB,yBAAyB,CACvC,QAAQ,EAAE,MAAM,EAChB,QAAQ,EAAE,MAAM,GACf,MAAM,CAER;AAED,wBAAgB,4BAA4B,CAC1C,OAAO,EAAE,MAAM,GACd,cAAc,GAAG,IAAI,CAEvB;AAED,wBAAgB,4BAA4B,CAC1C,OAAO,EAAE,MAAM,EACf,KAAK,EAAE,cAAc,GACpB,MAAM,CAkBR;AAED,wBAAgB,aAAa,CAAC,MAAM,EAAE,MAAM,CAAC,MAAM,EAAE,OAAO,CAAC,GAAG,MAAM,CAErE"}

package/dist/providers/codex/codec/shared.js

@@ -12,6 +12,10 @@ const STANDARD_SUPPORTED_CODEX_EFFORTS = [
     'xhigh',
 ];
 export const SUPPORTED_CODEX_ROLE_TARGETS = [
+    ...[...STANDARD_SUPPORTED_CODEX_EFFORTS, 'max'].map((effort) => ({
+        model: 'gpt-6-astra',
+        effort,
+    })),
     ...[...STANDARD_SUPPORTED_CODEX_EFFORTS, 'max'].map((effort) => ({
         model: 'gpt-6-luna',
         effort,

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@open-agent-toolkit/cli",
-  "version": "0.3.2",
+  "version": "0.3.3",
   "private": false,
   "description": "Open Agent Toolkit CLI",
   "homepage": "https://github.com/voxmedia/open-agent-toolkit/tree/main/packages/cli",
@@ -34,7 +34,7 @@
     "ora": "^9.0.0",
     "yaml": "2.8.2",
     "zod": "^3.25.76",
-    "@open-agent-toolkit/control-plane": "0.3.2"
+    "@open-agent-toolkit/control-plane": "0.3.3"
   },
   "devDependencies": {
     "@types/node": "^22.10.0",