@cubis/foundry 0.3.81 → 0.3.83

package/workflows/workflows/agent-environment-setup/platforms/codex/skills/code-review/SKILL.md

@@ -61,8 +61,15 @@ When conducting a code review, provide:
 
 ## Codex Platform Notes
 
-- Specialists are internal reasoning postures, not spawned subagent processes.
-- Reference the repo-root AGENTS instructions for posture definitions and switching contracts.
+- Codex supports native subagents via `.codex/agents/*.toml` files with `name`, `description`, and `developer_instructions`.
+- Each subagent TOML can specify `model` and `model_reasoning_effort` to optimize cost per task difficulty:
+  - Light tasks (exploration, docs): `model = "gpt-5.3-codex-spark"`, `model_reasoning_effort = "medium"`
+  - Heavy tasks (security audit, orchestration): `model = "gpt-5.4"`, `model_reasoning_effort = "high"`
+  - Standard tasks (implementation): inherit parent model (omit `model` field).
+- Built-in agents: `default`, `worker`, `explorer`. Custom agents extend these via TOML definitions.
 - Codex operates under network restrictions — skills should not assume outbound HTTP access.
 - Use `$ARGUMENTS` to access user-provided arguments when the skill is invoked.
 - All skill guidance executes within the sandbox; file I/O is confined to the workspace.
+- Skills are installed at `.agents/skills/<skill-id>/SKILL.md`. Workflow skills can also be compiled to `.agents/skills/<workflow-id>/SKILL.md` as `generatedSkills`.
+- Codex supports three autonomy levels: `suggest`, `auto-edit`, `full-auto`.
+- MCP skill tools are available when the Cubis Foundry MCP server is connected.

package/workflows/workflows/agent-environment-setup/platforms/codex/skills/csharp-best-practices/SKILL.md

@@ -76,8 +76,15 @@ No helper scripts are required for this skill right now. Keep execution in `SKIL
 
 ## Codex Platform Notes
 
-- Specialists are internal reasoning postures, not spawned subagent processes.
-- Reference the repo-root AGENTS instructions for posture definitions and switching contracts.
+- Codex supports native subagents via `.codex/agents/*.toml` files with `name`, `description`, and `developer_instructions`.
+- Each subagent TOML can specify `model` and `model_reasoning_effort` to optimize cost per task difficulty:
+  - Light tasks (exploration, docs): `model = "gpt-5.3-codex-spark"`, `model_reasoning_effort = "medium"`
+  - Heavy tasks (security audit, orchestration): `model = "gpt-5.4"`, `model_reasoning_effort = "high"`
+  - Standard tasks (implementation): inherit parent model (omit `model` field).
+- Built-in agents: `default`, `worker`, `explorer`. Custom agents extend these via TOML definitions.
 - Codex operates under network restrictions — skills should not assume outbound HTTP access.
 - Use `$ARGUMENTS` to access user-provided arguments when the skill is invoked.
 - All skill guidance executes within the sandbox; file I/O is confined to the workspace.
+- Skills are installed at `.agents/skills/<skill-id>/SKILL.md`. Workflow skills can also be compiled to `.agents/skills/<workflow-id>/SKILL.md` as `generatedSkills`.
+- Codex supports three autonomy levels: `suggest`, `auto-edit`, `full-auto`.
+- MCP skill tools are available when the Cubis Foundry MCP server is connected.

package/workflows/workflows/agent-environment-setup/platforms/codex/skills/database-design/SKILL.md

@@ -70,8 +70,15 @@ Load on demand. Do not preload all reference files.
 
 ## Codex Platform Notes
 
-- Specialists are internal reasoning postures, not spawned subagent processes.
-- Reference the repo-root AGENTS instructions for posture definitions and switching contracts.
+- Codex supports native subagents via `.codex/agents/*.toml` files with `name`, `description`, and `developer_instructions`.
+- Each subagent TOML can specify `model` and `model_reasoning_effort` to optimize cost per task difficulty:
+  - Light tasks (exploration, docs): `model = "gpt-5.3-codex-spark"`, `model_reasoning_effort = "medium"`
+  - Heavy tasks (security audit, orchestration): `model = "gpt-5.4"`, `model_reasoning_effort = "high"`
+  - Standard tasks (implementation): inherit parent model (omit `model` field).
+- Built-in agents: `default`, `worker`, `explorer`. Custom agents extend these via TOML definitions.
 - Codex operates under network restrictions — skills should not assume outbound HTTP access.
 - Use `$ARGUMENTS` to access user-provided arguments when the skill is invoked.
 - All skill guidance executes within the sandbox; file I/O is confined to the workspace.
+- Skills are installed at `.agents/skills/<skill-id>/SKILL.md`. Workflow skills can also be compiled to `.agents/skills/<workflow-id>/SKILL.md` as `generatedSkills`.
+- Codex supports three autonomy levels: `suggest`, `auto-edit`, `full-auto`.
+- MCP skill tools are available when the Cubis Foundry MCP server is connected.

package/workflows/workflows/agent-environment-setup/platforms/codex/skills/deep-research/SKILL.md

@@ -82,8 +82,15 @@ Use these when the task shape already matches.
 
 ## Codex Platform Notes
 
-- Specialists are internal reasoning postures, not spawned subagent processes.
-- Reference the repo-root AGENTS instructions for posture definitions and switching contracts.
+- Codex supports native subagents via `.codex/agents/*.toml` files with `name`, `description`, and `developer_instructions`.
+- Each subagent TOML can specify `model` and `model_reasoning_effort` to optimize cost per task difficulty:
+  - Light tasks (exploration, docs): `model = "gpt-5.3-codex-spark"`, `model_reasoning_effort = "medium"`
+  - Heavy tasks (security audit, orchestration): `model = "gpt-5.4"`, `model_reasoning_effort = "high"`
+  - Standard tasks (implementation): inherit parent model (omit `model` field).
+- Built-in agents: `default`, `worker`, `explorer`. Custom agents extend these via TOML definitions.
 - Codex operates under network restrictions — skills should not assume outbound HTTP access.
 - Use `$ARGUMENTS` to access user-provided arguments when the skill is invoked.
 - All skill guidance executes within the sandbox; file I/O is confined to the workspace.
+- Skills are installed at `.agents/skills/<skill-id>/SKILL.md`. Workflow skills can also be compiled to `.agents/skills/<workflow-id>/SKILL.md` as `generatedSkills`.
+- Codex supports three autonomy levels: `suggest`, `auto-edit`, `full-auto`.
+- MCP skill tools are available when the Cubis Foundry MCP server is connected.

package/workflows/workflows/agent-environment-setup/platforms/codex/skills/django-drf/SKILL.md

@@ -61,8 +61,15 @@ No helper scripts are required for this skill right now. Keep execution in `SKIL
 
 ## Codex Platform Notes
 
-- Specialists are internal reasoning postures, not spawned subagent processes.
-- Reference the repo-root AGENTS instructions for posture definitions and switching contracts.
+- Codex supports native subagents via `.codex/agents/*.toml` files with `name`, `description`, and `developer_instructions`.
+- Each subagent TOML can specify `model` and `model_reasoning_effort` to optimize cost per task difficulty:
+  - Light tasks (exploration, docs): `model = "gpt-5.3-codex-spark"`, `model_reasoning_effort = "medium"`
+  - Heavy tasks (security audit, orchestration): `model = "gpt-5.4"`, `model_reasoning_effort = "high"`
+  - Standard tasks (implementation): inherit parent model (omit `model` field).
+- Built-in agents: `default`, `worker`, `explorer`. Custom agents extend these via TOML definitions.
 - Codex operates under network restrictions — skills should not assume outbound HTTP access.
 - Use `$ARGUMENTS` to access user-provided arguments when the skill is invoked.
 - All skill guidance executes within the sandbox; file I/O is confined to the workspace.
+- Skills are installed at `.agents/skills/<skill-id>/SKILL.md`. Workflow skills can also be compiled to `.agents/skills/<workflow-id>/SKILL.md` as `generatedSkills`.
+- Codex supports three autonomy levels: `suggest`, `auto-edit`, `full-auto`.
+- MCP skill tools are available when the Cubis Foundry MCP server is connected.

package/workflows/workflows/agent-environment-setup/platforms/codex/skills/docker-compose-dev/SKILL.md

@@ -67,8 +67,15 @@ Provide complete `docker-compose.yml` and `Dockerfile` content with inline comme
 
 ## Codex Platform Notes
 
-- Specialists are internal reasoning postures, not spawned subagent processes.
-- Reference the repo-root AGENTS instructions for posture definitions and switching contracts.
+- Codex supports native subagents via `.codex/agents/*.toml` files with `name`, `description`, and `developer_instructions`.
+- Each subagent TOML can specify `model` and `model_reasoning_effort` to optimize cost per task difficulty:
+  - Light tasks (exploration, docs): `model = "gpt-5.3-codex-spark"`, `model_reasoning_effort = "medium"`
+  - Heavy tasks (security audit, orchestration): `model = "gpt-5.4"`, `model_reasoning_effort = "high"`
+  - Standard tasks (implementation): inherit parent model (omit `model` field).
+- Built-in agents: `default`, `worker`, `explorer`. Custom agents extend these via TOML definitions.
 - Codex operates under network restrictions — skills should not assume outbound HTTP access.
 - Use `$ARGUMENTS` to access user-provided arguments when the skill is invoked.
 - All skill guidance executes within the sandbox; file I/O is confined to the workspace.
+- Skills are installed at `.agents/skills/<skill-id>/SKILL.md`. Workflow skills can also be compiled to `.agents/skills/<workflow-id>/SKILL.md` as `generatedSkills`.
+- Codex supports three autonomy levels: `suggest`, `auto-edit`, `full-auto`.
+- MCP skill tools are available when the Cubis Foundry MCP server is connected.

package/workflows/workflows/agent-environment-setup/platforms/codex/skills/drizzle-orm/SKILL.md

@@ -74,8 +74,15 @@ No helper scripts are required for this skill right now. Keep execution in `SKIL
 
 ## Codex Platform Notes
 
-- Specialists are internal reasoning postures, not spawned subagent processes.
-- Reference the repo-root AGENTS instructions for posture definitions and switching contracts.
+- Codex supports native subagents via `.codex/agents/*.toml` files with `name`, `description`, and `developer_instructions`.
+- Each subagent TOML can specify `model` and `model_reasoning_effort` to optimize cost per task difficulty:
+  - Light tasks (exploration, docs): `model = "gpt-5.3-codex-spark"`, `model_reasoning_effort = "medium"`
+  - Heavy tasks (security audit, orchestration): `model = "gpt-5.4"`, `model_reasoning_effort = "high"`
+  - Standard tasks (implementation): inherit parent model (omit `model` field).
+- Built-in agents: `default`, `worker`, `explorer`. Custom agents extend these via TOML definitions.
 - Codex operates under network restrictions — skills should not assume outbound HTTP access.
 - Use `$ARGUMENTS` to access user-provided arguments when the skill is invoked.
 - All skill guidance executes within the sandbox; file I/O is confined to the workspace.
+- Skills are installed at `.agents/skills/<skill-id>/SKILL.md`. Workflow skills can also be compiled to `.agents/skills/<workflow-id>/SKILL.md` as `generatedSkills`.
+- Codex supports three autonomy levels: `suggest`, `auto-edit`, `full-auto`.
+- MCP skill tools are available when the Cubis Foundry MCP server is connected.

package/workflows/workflows/agent-environment-setup/platforms/codex/skills/electron-qa/SKILL.md

@@ -80,8 +80,15 @@ Deliver:
 
 ## Codex Platform Notes
 
-- Specialists are internal reasoning postures, not spawned subagent processes.
-- Reference the repo-root AGENTS instructions for posture definitions and switching contracts.
+- Codex supports native subagents via `.codex/agents/*.toml` files with `name`, `description`, and `developer_instructions`.
+- Each subagent TOML can specify `model` and `model_reasoning_effort` to optimize cost per task difficulty:
+  - Light tasks (exploration, docs): `model = "gpt-5.3-codex-spark"`, `model_reasoning_effort = "medium"`
+  - Heavy tasks (security audit, orchestration): `model = "gpt-5.4"`, `model_reasoning_effort = "high"`
+  - Standard tasks (implementation): inherit parent model (omit `model` field).
+- Built-in agents: `default`, `worker`, `explorer`. Custom agents extend these via TOML definitions.
 - Codex operates under network restrictions — skills should not assume outbound HTTP access.
 - Use `$ARGUMENTS` to access user-provided arguments when the skill is invoked.
 - All skill guidance executes within the sandbox; file I/O is confined to the workspace.
+- Skills are installed at `.agents/skills/<skill-id>/SKILL.md`. Workflow skills can also be compiled to `.agents/skills/<workflow-id>/SKILL.md` as `generatedSkills`.
+- Codex supports three autonomy levels: `suggest`, `auto-edit`, `full-auto`.
+- MCP skill tools are available when the Cubis Foundry MCP server is connected.

package/workflows/workflows/agent-environment-setup/platforms/codex/skills/expo-app/SKILL.md

@@ -86,8 +86,15 @@ Guide the design and implementation of production-grade Expo applications using
 
 ## Codex Platform Notes
 
-- Specialists are internal reasoning postures, not spawned subagent processes.
-- Reference the repo-root AGENTS instructions for posture definitions and switching contracts.
+- Codex supports native subagents via `.codex/agents/*.toml` files with `name`, `description`, and `developer_instructions`.
+- Each subagent TOML can specify `model` and `model_reasoning_effort` to optimize cost per task difficulty:
+  - Light tasks (exploration, docs): `model = "gpt-5.3-codex-spark"`, `model_reasoning_effort = "medium"`
+  - Heavy tasks (security audit, orchestration): `model = "gpt-5.4"`, `model_reasoning_effort = "high"`
+  - Standard tasks (implementation): inherit parent model (omit `model` field).
+- Built-in agents: `default`, `worker`, `explorer`. Custom agents extend these via TOML definitions.
 - Codex operates under network restrictions — skills should not assume outbound HTTP access.
 - Use `$ARGUMENTS` to access user-provided arguments when the skill is invoked.
 - All skill guidance executes within the sandbox; file I/O is confined to the workspace.
+- Skills are installed at `.agents/skills/<skill-id>/SKILL.md`. Workflow skills can also be compiled to `.agents/skills/<workflow-id>/SKILL.md` as `generatedSkills`.
+- Codex supports three autonomy levels: `suggest`, `auto-edit`, `full-auto`.
+- MCP skill tools are available when the Cubis Foundry MCP server is connected.

package/workflows/workflows/agent-environment-setup/platforms/codex/skills/express-nodejs/SKILL.md

@@ -73,8 +73,15 @@ Load only what the current step needs.
 
 ## Codex Platform Notes
 
-- Specialists are internal reasoning postures, not spawned subagent processes.
-- Reference the repo-root AGENTS instructions for posture definitions and switching contracts.
+- Codex supports native subagents via `.codex/agents/*.toml` files with `name`, `description`, and `developer_instructions`.
+- Each subagent TOML can specify `model` and `model_reasoning_effort` to optimize cost per task difficulty:
+  - Light tasks (exploration, docs): `model = "gpt-5.3-codex-spark"`, `model_reasoning_effort = "medium"`
+  - Heavy tasks (security audit, orchestration): `model = "gpt-5.4"`, `model_reasoning_effort = "high"`
+  - Standard tasks (implementation): inherit parent model (omit `model` field).
+- Built-in agents: `default`, `worker`, `explorer`. Custom agents extend these via TOML definitions.
 - Codex operates under network restrictions — skills should not assume outbound HTTP access.
 - Use `$ARGUMENTS` to access user-provided arguments when the skill is invoked.
 - All skill guidance executes within the sandbox; file I/O is confined to the workspace.
+- Skills are installed at `.agents/skills/<skill-id>/SKILL.md`. Workflow skills can also be compiled to `.agents/skills/<workflow-id>/SKILL.md` as `generatedSkills`.
+- Codex supports three autonomy levels: `suggest`, `auto-edit`, `full-auto`.
+- MCP skill tools are available when the Cubis Foundry MCP server is connected.

package/workflows/workflows/agent-environment-setup/platforms/codex/skills/fastapi/SKILL.md

@@ -67,8 +67,15 @@ Provide implementation code, Pydantic model definitions, dependency functions, c
 
 ## Codex Platform Notes
 
-- Specialists are internal reasoning postures, not spawned subagent processes.
-- Reference the repo-root AGENTS instructions for posture definitions and switching contracts.
+- Codex supports native subagents via `.codex/agents/*.toml` files with `name`, `description`, and `developer_instructions`.
+- Each subagent TOML can specify `model` and `model_reasoning_effort` to optimize cost per task difficulty:
+  - Light tasks (exploration, docs): `model = "gpt-5.3-codex-spark"`, `model_reasoning_effort = "medium"`
+  - Heavy tasks (security audit, orchestration): `model = "gpt-5.4"`, `model_reasoning_effort = "high"`
+  - Standard tasks (implementation): inherit parent model (omit `model` field).
+- Built-in agents: `default`, `worker`, `explorer`. Custom agents extend these via TOML definitions.
 - Codex operates under network restrictions — skills should not assume outbound HTTP access.
 - Use `$ARGUMENTS` to access user-provided arguments when the skill is invoked.
 - All skill guidance executes within the sandbox; file I/O is confined to the workspace.
+- Skills are installed at `.agents/skills/<skill-id>/SKILL.md`. Workflow skills can also be compiled to `.agents/skills/<workflow-id>/SKILL.md` as `generatedSkills`.
+- Codex supports three autonomy levels: `suggest`, `auto-edit`, `full-auto`.
+- MCP skill tools are available when the Cubis Foundry MCP server is connected.

package/workflows/workflows/agent-environment-setup/platforms/codex/skills/frontend-design/SKILL.md

@@ -84,8 +84,15 @@ Load only what the current task requires.
 
 ## Codex Platform Notes
 
-- Specialists are internal reasoning postures, not spawned subagent processes.
-- Reference the repo-root AGENTS instructions for posture definitions and switching contracts.
+- Codex supports native subagents via `.codex/agents/*.toml` files with `name`, `description`, and `developer_instructions`.
+- Each subagent TOML can specify `model` and `model_reasoning_effort` to optimize cost per task difficulty:
+  - Light tasks (exploration, docs): `model = "gpt-5.3-codex-spark"`, `model_reasoning_effort = "medium"`
+  - Heavy tasks (security audit, orchestration): `model = "gpt-5.4"`, `model_reasoning_effort = "high"`
+  - Standard tasks (implementation): inherit parent model (omit `model` field).
+- Built-in agents: `default`, `worker`, `explorer`. Custom agents extend these via TOML definitions.
 - Codex operates under network restrictions — skills should not assume outbound HTTP access.
 - Use `$ARGUMENTS` to access user-provided arguments when the skill is invoked.
 - All skill guidance executes within the sandbox; file I/O is confined to the workspace.
+- Skills are installed at `.agents/skills/<skill-id>/SKILL.md`. Workflow skills can also be compiled to `.agents/skills/<workflow-id>/SKILL.md` as `generatedSkills`.
+- Codex supports three autonomy levels: `suggest`, `auto-edit`, `full-auto`.
+- MCP skill tools are available when the Cubis Foundry MCP server is connected.

package/workflows/workflows/agent-environment-setup/platforms/codex/skills/gin-golang/SKILL.md

@@ -69,8 +69,15 @@ Load only what the current step needs.
 
 ## Codex Platform Notes
 
-- Specialists are internal reasoning postures, not spawned subagent processes.
-- Reference the repo-root AGENTS instructions for posture definitions and switching contracts.
+- Codex supports native subagents via `.codex/agents/*.toml` files with `name`, `description`, and `developer_instructions`.
+- Each subagent TOML can specify `model` and `model_reasoning_effort` to optimize cost per task difficulty:
+  - Light tasks (exploration, docs): `model = "gpt-5.3-codex-spark"`, `model_reasoning_effort = "medium"`
+  - Heavy tasks (security audit, orchestration): `model = "gpt-5.4"`, `model_reasoning_effort = "high"`
+  - Standard tasks (implementation): inherit parent model (omit `model` field).
+- Built-in agents: `default`, `worker`, `explorer`. Custom agents extend these via TOML definitions.
 - Codex operates under network restrictions — skills should not assume outbound HTTP access.
 - Use `$ARGUMENTS` to access user-provided arguments when the skill is invoked.
 - All skill guidance executes within the sandbox; file I/O is confined to the workspace.
+- Skills are installed at `.agents/skills/<skill-id>/SKILL.md`. Workflow skills can also be compiled to `.agents/skills/<workflow-id>/SKILL.md` as `generatedSkills`.
+- Codex supports three autonomy levels: `suggest`, `auto-edit`, `full-auto`.
+- MCP skill tools are available when the Cubis Foundry MCP server is connected.

package/workflows/workflows/agent-environment-setup/platforms/codex/skills/git-workflow/SKILL.md

@@ -61,8 +61,15 @@ When advising on git workflows, provide:
 
 ## Codex Platform Notes
 
-- Specialists are internal reasoning postures, not spawned subagent processes.
-- Reference the repo-root AGENTS instructions for posture definitions and switching contracts.
+- Codex supports native subagents via `.codex/agents/*.toml` files with `name`, `description`, and `developer_instructions`.
+- Each subagent TOML can specify `model` and `model_reasoning_effort` to optimize cost per task difficulty:
+  - Light tasks (exploration, docs): `model = "gpt-5.3-codex-spark"`, `model_reasoning_effort = "medium"`
+  - Heavy tasks (security audit, orchestration): `model = "gpt-5.4"`, `model_reasoning_effort = "high"`
+  - Standard tasks (implementation): inherit parent model (omit `model` field).
+- Built-in agents: `default`, `worker`, `explorer`. Custom agents extend these via TOML definitions.
 - Codex operates under network restrictions — skills should not assume outbound HTTP access.
 - Use `$ARGUMENTS` to access user-provided arguments when the skill is invoked.
 - All skill guidance executes within the sandbox; file I/O is confined to the workspace.
+- Skills are installed at `.agents/skills/<skill-id>/SKILL.md`. Workflow skills can also be compiled to `.agents/skills/<workflow-id>/SKILL.md` as `generatedSkills`.
+- Codex supports three autonomy levels: `suggest`, `auto-edit`, `full-auto`.
+- MCP skill tools are available when the Cubis Foundry MCP server is connected.

package/workflows/workflows/agent-environment-setup/platforms/codex/skills/go-fiber/SKILL.md

@@ -69,8 +69,15 @@ Load only what the current step needs.
 
 ## Codex Platform Notes
 
-- Specialists are internal reasoning postures, not spawned subagent processes.
-- Reference the repo-root AGENTS instructions for posture definitions and switching contracts.
+- Codex supports native subagents via `.codex/agents/*.toml` files with `name`, `description`, and `developer_instructions`.
+- Each subagent TOML can specify `model` and `model_reasoning_effort` to optimize cost per task difficulty:
+  - Light tasks (exploration, docs): `model = "gpt-5.3-codex-spark"`, `model_reasoning_effort = "medium"`
+  - Heavy tasks (security audit, orchestration): `model = "gpt-5.4"`, `model_reasoning_effort = "high"`
+  - Standard tasks (implementation): inherit parent model (omit `model` field).
+- Built-in agents: `default`, `worker`, `explorer`. Custom agents extend these via TOML definitions.
 - Codex operates under network restrictions — skills should not assume outbound HTTP access.
 - Use `$ARGUMENTS` to access user-provided arguments when the skill is invoked.
 - All skill guidance executes within the sandbox; file I/O is confined to the workspace.
+- Skills are installed at `.agents/skills/<skill-id>/SKILL.md`. Workflow skills can also be compiled to `.agents/skills/<workflow-id>/SKILL.md` as `generatedSkills`.
+- Codex supports three autonomy levels: `suggest`, `auto-edit`, `full-auto`.
+- MCP skill tools are available when the Cubis Foundry MCP server is connected.

package/workflows/workflows/agent-environment-setup/platforms/codex/skills/golang-best-practices/SKILL.md

@@ -72,8 +72,15 @@ No helper scripts are required for this skill right now. Keep execution in `SKIL
 
 ## Codex Platform Notes
 
-- Specialists are internal reasoning postures, not spawned subagent processes.
-- Reference the repo-root AGENTS instructions for posture definitions and switching contracts.
+- Codex supports native subagents via `.codex/agents/*.toml` files with `name`, `description`, and `developer_instructions`.
+- Each subagent TOML can specify `model` and `model_reasoning_effort` to optimize cost per task difficulty:
+  - Light tasks (exploration, docs): `model = "gpt-5.3-codex-spark"`, `model_reasoning_effort = "medium"`
+  - Heavy tasks (security audit, orchestration): `model = "gpt-5.4"`, `model_reasoning_effort = "high"`
+  - Standard tasks (implementation): inherit parent model (omit `model` field).
+- Built-in agents: `default`, `worker`, `explorer`. Custom agents extend these via TOML definitions.
 - Codex operates under network restrictions — skills should not assume outbound HTTP access.
 - Use `$ARGUMENTS` to access user-provided arguments when the skill is invoked.
 - All skill guidance executes within the sandbox; file I/O is confined to the workspace.
+- Skills are installed at `.agents/skills/<skill-id>/SKILL.md`. Workflow skills can also be compiled to `.agents/skills/<workflow-id>/SKILL.md` as `generatedSkills`.
+- Codex supports three autonomy levels: `suggest`, `auto-edit`, `full-auto`.
+- MCP skill tools are available when the Cubis Foundry MCP server is connected.

package/workflows/workflows/agent-environment-setup/platforms/codex/skills/google-workspace/SKILL.md

@@ -84,8 +84,15 @@ Guide the design and implementation of production-grade Google Workspace API int
 
 ## Codex Platform Notes
 
-- Specialists are internal reasoning postures, not spawned subagent processes.
-- Reference the repo-root AGENTS instructions for posture definitions and switching contracts.
+- Codex supports native subagents via `.codex/agents/*.toml` files with `name`, `description`, and `developer_instructions`.
+- Each subagent TOML can specify `model` and `model_reasoning_effort` to optimize cost per task difficulty:
+  - Light tasks (exploration, docs): `model = "gpt-5.3-codex-spark"`, `model_reasoning_effort = "medium"`
+  - Heavy tasks (security audit, orchestration): `model = "gpt-5.4"`, `model_reasoning_effort = "high"`
+  - Standard tasks (implementation): inherit parent model (omit `model` field).
+- Built-in agents: `default`, `worker`, `explorer`. Custom agents extend these via TOML definitions.
 - Codex operates under network restrictions — skills should not assume outbound HTTP access.
 - Use `$ARGUMENTS` to access user-provided arguments when the skill is invoked.
 - All skill guidance executes within the sandbox; file I/O is confined to the workspace.
+- Skills are installed at `.agents/skills/<skill-id>/SKILL.md`. Workflow skills can also be compiled to `.agents/skills/<workflow-id>/SKILL.md` as `generatedSkills`.
+- Codex supports three autonomy levels: `suggest`, `auto-edit`, `full-auto`.
+- MCP skill tools are available when the Cubis Foundry MCP server is connected.

package/workflows/workflows/agent-environment-setup/platforms/codex/skills/huggingface-ml/SKILL.md

@@ -83,8 +83,15 @@ Guide the design and implementation of machine learning workflows using the Hugg
 
 ## Codex Platform Notes
 
-- Specialists are internal reasoning postures, not spawned subagent processes.
-- Reference the repo-root AGENTS instructions for posture definitions and switching contracts.
+- Codex supports native subagents via `.codex/agents/*.toml` files with `name`, `description`, and `developer_instructions`.
+- Each subagent TOML can specify `model` and `model_reasoning_effort` to optimize cost per task difficulty:
+  - Light tasks (exploration, docs): `model = "gpt-5.3-codex-spark"`, `model_reasoning_effort = "medium"`
+  - Heavy tasks (security audit, orchestration): `model = "gpt-5.4"`, `model_reasoning_effort = "high"`
+  - Standard tasks (implementation): inherit parent model (omit `model` field).
+- Built-in agents: `default`, `worker`, `explorer`. Custom agents extend these via TOML definitions.
 - Codex operates under network restrictions — skills should not assume outbound HTTP access.
 - Use `$ARGUMENTS` to access user-provided arguments when the skill is invoked.
 - All skill guidance executes within the sandbox; file I/O is confined to the workspace.
+- Skills are installed at `.agents/skills/<skill-id>/SKILL.md`. Workflow skills can also be compiled to `.agents/skills/<workflow-id>/SKILL.md` as `generatedSkills`.
+- Codex supports three autonomy levels: `suggest`, `auto-edit`, `full-auto`.
+- MCP skill tools are available when the Cubis Foundry MCP server is connected.

package/workflows/workflows/agent-environment-setup/platforms/codex/skills/integration-testing/SKILL.md

@@ -98,8 +98,15 @@ Provide a structured methodology for writing integration tests that verify corre
 
 ## Codex Platform Notes
 
-- Specialists are internal reasoning postures, not spawned subagent processes.
-- Reference the repo-root AGENTS instructions for posture definitions and switching contracts.
+- Codex supports native subagents via `.codex/agents/*.toml` files with `name`, `description`, and `developer_instructions`.
+- Each subagent TOML can specify `model` and `model_reasoning_effort` to optimize cost per task difficulty:
+  - Light tasks (exploration, docs): `model = "gpt-5.3-codex-spark"`, `model_reasoning_effort = "medium"`
+  - Heavy tasks (security audit, orchestration): `model = "gpt-5.4"`, `model_reasoning_effort = "high"`
+  - Standard tasks (implementation): inherit parent model (omit `model` field).
+- Built-in agents: `default`, `worker`, `explorer`. Custom agents extend these via TOML definitions.
 - Codex operates under network restrictions — skills should not assume outbound HTTP access.
 - Use `$ARGUMENTS` to access user-provided arguments when the skill is invoked.
 - All skill guidance executes within the sandbox; file I/O is confined to the workspace.
+- Skills are installed at `.agents/skills/<skill-id>/SKILL.md`. Workflow skills can also be compiled to `.agents/skills/<workflow-id>/SKILL.md` as `generatedSkills`.
+- Codex supports three autonomy levels: `suggest`, `auto-edit`, `full-auto`.
+- MCP skill tools are available when the Cubis Foundry MCP server is connected.

package/workflows/workflows/agent-environment-setup/platforms/codex/skills/java-best-practices/SKILL.md

@@ -67,8 +67,15 @@ Produces Java code using records, sealed types, pattern matching switch expressi
 
 ## Codex Platform Notes
 
-- Specialists are internal reasoning postures, not spawned subagent processes.
-- Reference the repo-root AGENTS instructions for posture definitions and switching contracts.
+- Codex supports native subagents via `.codex/agents/*.toml` files with `name`, `description`, and `developer_instructions`.
+- Each subagent TOML can specify `model` and `model_reasoning_effort` to optimize cost per task difficulty:
+  - Light tasks (exploration, docs): `model = "gpt-5.3-codex-spark"`, `model_reasoning_effort = "medium"`
+  - Heavy tasks (security audit, orchestration): `model = "gpt-5.4"`, `model_reasoning_effort = "high"`
+  - Standard tasks (implementation): inherit parent model (omit `model` field).
+- Built-in agents: `default`, `worker`, `explorer`. Custom agents extend these via TOML definitions.
 - Codex operates under network restrictions — skills should not assume outbound HTTP access.
 - Use `$ARGUMENTS` to access user-provided arguments when the skill is invoked.
 - All skill guidance executes within the sandbox; file I/O is confined to the workspace.
+- Skills are installed at `.agents/skills/<skill-id>/SKILL.md`. Workflow skills can also be compiled to `.agents/skills/<workflow-id>/SKILL.md` as `generatedSkills`.
+- Codex supports three autonomy levels: `suggest`, `auto-edit`, `full-auto`.
+- MCP skill tools are available when the Cubis Foundry MCP server is connected.

package/workflows/workflows/agent-environment-setup/platforms/codex/skills/javascript-best-practices/SKILL.md

@@ -67,8 +67,15 @@ Produces JavaScript code using ESM modules, native runtime APIs, explicit async
 
 ## Codex Platform Notes
 
-- Specialists are internal reasoning postures, not spawned subagent processes.
-- Reference the repo-root AGENTS instructions for posture definitions and switching contracts.
+- Codex supports native subagents via `.codex/agents/*.toml` files with `name`, `description`, and `developer_instructions`.
+- Each subagent TOML can specify `model` and `model_reasoning_effort` to optimize cost per task difficulty:
+  - Light tasks (exploration, docs): `model = "gpt-5.3-codex-spark"`, `model_reasoning_effort = "medium"`
+  - Heavy tasks (security audit, orchestration): `model = "gpt-5.4"`, `model_reasoning_effort = "high"`
+  - Standard tasks (implementation): inherit parent model (omit `model` field).
+- Built-in agents: `default`, `worker`, `explorer`. Custom agents extend these via TOML definitions.
 - Codex operates under network restrictions — skills should not assume outbound HTTP access.
 - Use `$ARGUMENTS` to access user-provided arguments when the skill is invoked.
 - All skill guidance executes within the sandbox; file I/O is confined to the workspace.
+- Skills are installed at `.agents/skills/<skill-id>/SKILL.md`. Workflow skills can also be compiled to `.agents/skills/<workflow-id>/SKILL.md` as `generatedSkills`.
+- Codex supports three autonomy levels: `suggest`, `auto-edit`, `full-auto`.
+- MCP skill tools are available when the Cubis Foundry MCP server is connected.

package/workflows/workflows/agent-environment-setup/platforms/codex/skills/kaizen-iteration/SKILL.md

@@ -93,8 +93,15 @@ Provide a structured, metrics-driven framework for continuous improvement in sof
 
 ## Codex Platform Notes
 
-- Specialists are internal reasoning postures, not spawned subagent processes.
-- Reference the repo-root AGENTS instructions for posture definitions and switching contracts.
+- Codex supports native subagents via `.codex/agents/*.toml` files with `name`, `description`, and `developer_instructions`.
+- Each subagent TOML can specify `model` and `model_reasoning_effort` to optimize cost per task difficulty:
+  - Light tasks (exploration, docs): `model = "gpt-5.3-codex-spark"`, `model_reasoning_effort = "medium"`
+  - Heavy tasks (security audit, orchestration): `model = "gpt-5.4"`, `model_reasoning_effort = "high"`
+  - Standard tasks (implementation): inherit parent model (omit `model` field).
+- Built-in agents: `default`, `worker`, `explorer`. Custom agents extend these via TOML definitions.
 - Codex operates under network restrictions — skills should not assume outbound HTTP access.
 - Use `$ARGUMENTS` to access user-provided arguments when the skill is invoked.
 - All skill guidance executes within the sandbox; file I/O is confined to the workspace.
+- Skills are installed at `.agents/skills/<skill-id>/SKILL.md`. Workflow skills can also be compiled to `.agents/skills/<workflow-id>/SKILL.md` as `generatedSkills`.
+- Codex supports three autonomy levels: `suggest`, `auto-edit`, `full-auto`.
+- MCP skill tools are available when the Cubis Foundry MCP server is connected.

package/workflows/workflows/agent-environment-setup/platforms/codex/skills/kotlin-best-practices/SKILL.md

@@ -67,8 +67,15 @@ Produces Kotlin code using data classes, sealed hierarchies, structured coroutin
 
 ## Codex Platform Notes
 
-- Specialists are internal reasoning postures, not spawned subagent processes.
-- Reference the repo-root AGENTS instructions for posture definitions and switching contracts.
+- Codex supports native subagents via `.codex/agents/*.toml` files with `name`, `description`, and `developer_instructions`.
+- Each subagent TOML can specify `model` and `model_reasoning_effort` to optimize cost per task difficulty:
+  - Light tasks (exploration, docs): `model = "gpt-5.3-codex-spark"`, `model_reasoning_effort = "medium"`
+  - Heavy tasks (security audit, orchestration): `model = "gpt-5.4"`, `model_reasoning_effort = "high"`
+  - Standard tasks (implementation): inherit parent model (omit `model` field).
+- Built-in agents: `default`, `worker`, `explorer`. Custom agents extend these via TOML definitions.
 - Codex operates under network restrictions — skills should not assume outbound HTTP access.
 - Use `$ARGUMENTS` to access user-provided arguments when the skill is invoked.
 - All skill guidance executes within the sandbox; file I/O is confined to the workspace.
+- Skills are installed at `.agents/skills/<skill-id>/SKILL.md`. Workflow skills can also be compiled to `.agents/skills/<workflow-id>/SKILL.md` as `generatedSkills`.
+- Codex supports three autonomy levels: `suggest`, `auto-edit`, `full-auto`.
+- MCP skill tools are available when the Cubis Foundry MCP server is connected.

package/workflows/workflows/agent-environment-setup/platforms/codex/skills/kubernetes-deploy/SKILL.md

@@ -71,8 +71,15 @@ Provide complete Kubernetes manifests in YAML with inline comments explaining no
 
 ## Codex Platform Notes
 
-- Specialists are internal reasoning postures, not spawned subagent processes.
-- Reference the repo-root AGENTS instructions for posture definitions and switching contracts.
+- Codex supports native subagents via `.codex/agents/*.toml` files with `name`, `description`, and `developer_instructions`.
+- Each subagent TOML can specify `model` and `model_reasoning_effort` to optimize cost per task difficulty:
+  - Light tasks (exploration, docs): `model = "gpt-5.3-codex-spark"`, `model_reasoning_effort = "medium"`
+  - Heavy tasks (security audit, orchestration): `model = "gpt-5.4"`, `model_reasoning_effort = "high"`
+  - Standard tasks (implementation): inherit parent model (omit `model` field).
+- Built-in agents: `default`, `worker`, `explorer`. Custom agents extend these via TOML definitions.
 - Codex operates under network restrictions — skills should not assume outbound HTTP access.
 - Use `$ARGUMENTS` to access user-provided arguments when the skill is invoked.
 - All skill guidance executes within the sandbox; file I/O is confined to the workspace.
+- Skills are installed at `.agents/skills/<skill-id>/SKILL.md`. Workflow skills can also be compiled to `.agents/skills/<workflow-id>/SKILL.md` as `generatedSkills`.
+- Codex supports three autonomy levels: `suggest`, `auto-edit`, `full-auto`.
+- MCP skill tools are available when the Cubis Foundry MCP server is connected.

package/workflows/workflows/agent-environment-setup/platforms/codex/skills/laravel/SKILL.md

@@ -61,8 +61,15 @@ No helper scripts are required for this skill right now. Keep execution in `SKIL
 
 ## Codex Platform Notes
 
-- Specialists are internal reasoning postures, not spawned subagent processes.
-- Reference the repo-root AGENTS instructions for posture definitions and switching contracts.
+- Codex supports native subagents via `.codex/agents/*.toml` files with `name`, `description`, and `developer_instructions`.
+- Each subagent TOML can specify `model` and `model_reasoning_effort` to optimize cost per task difficulty:
+  - Light tasks (exploration, docs): `model = "gpt-5.3-codex-spark"`, `model_reasoning_effort = "medium"`
+  - Heavy tasks (security audit, orchestration): `model = "gpt-5.4"`, `model_reasoning_effort = "high"`
+  - Standard tasks (implementation): inherit parent model (omit `model` field).
+- Built-in agents: `default`, `worker`, `explorer`. Custom agents extend these via TOML definitions.
 - Codex operates under network restrictions — skills should not assume outbound HTTP access.
 - Use `$ARGUMENTS` to access user-provided arguments when the skill is invoked.
 - All skill guidance executes within the sandbox; file I/O is confined to the workspace.
+- Skills are installed at `.agents/skills/<skill-id>/SKILL.md`. Workflow skills can also be compiled to `.agents/skills/<workflow-id>/SKILL.md` as `generatedSkills`.
+- Codex supports three autonomy levels: `suggest`, `auto-edit`, `full-auto`.
+- MCP skill tools are available when the Cubis Foundry MCP server is connected.

package/workflows/workflows/agent-environment-setup/platforms/codex/skills/llm-eval/SKILL.md

@@ -67,8 +67,15 @@ Provide eval suite designs as structured JSON or YAML with assertion definitions
 
 ## Codex Platform Notes
 
-- Specialists are internal reasoning postures, not spawned subagent processes.
-- Reference the repo-root AGENTS instructions for posture definitions and switching contracts.
+- Codex supports native subagents via `.codex/agents/*.toml` files with `name`, `description`, and `developer_instructions`.
+- Each subagent TOML can specify `model` and `model_reasoning_effort` to optimize cost per task difficulty:
+  - Light tasks (exploration, docs): `model = "gpt-5.3-codex-spark"`, `model_reasoning_effort = "medium"`
+  - Heavy tasks (security audit, orchestration): `model = "gpt-5.4"`, `model_reasoning_effort = "high"`
+  - Standard tasks (implementation): inherit parent model (omit `model` field).
+- Built-in agents: `default`, `worker`, `explorer`. Custom agents extend these via TOML definitions.
 - Codex operates under network restrictions — skills should not assume outbound HTTP access.
 - Use `$ARGUMENTS` to access user-provided arguments when the skill is invoked.
 - All skill guidance executes within the sandbox; file I/O is confined to the workspace.
+- Skills are installed at `.agents/skills/<skill-id>/SKILL.md`. Workflow skills can also be compiled to `.agents/skills/<workflow-id>/SKILL.md` as `generatedSkills`.
+- Codex supports three autonomy levels: `suggest`, `auto-edit`, `full-auto`.
+- MCP skill tools are available when the Cubis Foundry MCP server is connected.

package/workflows/workflows/agent-environment-setup/platforms/codex/skills/mcp-server-builder/SKILL.md

@@ -83,8 +83,15 @@ Guide the design and implementation of production-grade Model Context Protocol (
 
 ## Codex Platform Notes
 
-- Specialists are internal reasoning postures, not spawned subagent processes.
-- Reference the repo-root AGENTS instructions for posture definitions and switching contracts.
+- Codex supports native subagents via `.codex/agents/*.toml` files with `name`, `description`, and `developer_instructions`.
+- Each subagent TOML can specify `model` and `model_reasoning_effort` to optimize cost per task difficulty:
+  - Light tasks (exploration, docs): `model = "gpt-5.3-codex-spark"`, `model_reasoning_effort = "medium"`
+  - Heavy tasks (security audit, orchestration): `model = "gpt-5.4"`, `model_reasoning_effort = "high"`
+  - Standard tasks (implementation): inherit parent model (omit `model` field).
+- Built-in agents: `default`, `worker`, `explorer`. Custom agents extend these via TOML definitions.
 - Codex operates under network restrictions — skills should not assume outbound HTTP access.
 - Use `$ARGUMENTS` to access user-provided arguments when the skill is invoked.
 - All skill guidance executes within the sandbox; file I/O is confined to the workspace.
+- Skills are installed at `.agents/skills/<skill-id>/SKILL.md`. Workflow skills can also be compiled to `.agents/skills/<workflow-id>/SKILL.md` as `generatedSkills`.
+- Codex supports three autonomy levels: `suggest`, `auto-edit`, `full-auto`.
+- MCP skill tools are available when the Cubis Foundry MCP server is connected.

package/workflows/workflows/agent-environment-setup/platforms/codex/skills/microservices-design/SKILL.md

@@ -77,8 +77,15 @@ Load only what the current task requires.
 
 ## Codex Platform Notes
 
-- Specialists are internal reasoning postures, not spawned subagent processes.
-- Reference the repo-root AGENTS instructions for posture definitions and switching contracts.
+- Codex supports native subagents via `.codex/agents/*.toml` files with `name`, `description`, and `developer_instructions`.
+- Each subagent TOML can specify `model` and `model_reasoning_effort` to optimize cost per task difficulty:
+  - Light tasks (exploration, docs): `model = "gpt-5.3-codex-spark"`, `model_reasoning_effort = "medium"`
+  - Heavy tasks (security audit, orchestration): `model = "gpt-5.4"`, `model_reasoning_effort = "high"`
+  - Standard tasks (implementation): inherit parent model (omit `model` field).
+- Built-in agents: `default`, `worker`, `explorer`. Custom agents extend these via TOML definitions.
 - Codex operates under network restrictions — skills should not assume outbound HTTP access.
 - Use `$ARGUMENTS` to access user-provided arguments when the skill is invoked.
 - All skill guidance executes within the sandbox; file I/O is confined to the workspace.
+- Skills are installed at `.agents/skills/<skill-id>/SKILL.md`. Workflow skills can also be compiled to `.agents/skills/<workflow-id>/SKILL.md` as `generatedSkills`.
+- Codex supports three autonomy levels: `suggest`, `auto-edit`, `full-auto`.
+- MCP skill tools are available when the Cubis Foundry MCP server is connected.

package/workflows/workflows/agent-environment-setup/platforms/codex/skills/nestjs/SKILL.md

@@ -67,8 +67,15 @@ Provide implementation code, module definitions, decorator usage, configuration
 
 ## Codex Platform Notes
 
-- Specialists are internal reasoning postures, not spawned subagent processes.
-- Reference the repo-root AGENTS instructions for posture definitions and switching contracts.
+- Codex supports native subagents via `.codex/agents/*.toml` files with `name`, `description`, and `developer_instructions`.
+- Each subagent TOML can specify `model` and `model_reasoning_effort` to optimize cost per task difficulty:
+  - Light tasks (exploration, docs): `model = "gpt-5.3-codex-spark"`, `model_reasoning_effort = "medium"`
+  - Heavy tasks (security audit, orchestration): `model = "gpt-5.4"`, `model_reasoning_effort = "high"`
+  - Standard tasks (implementation): inherit parent model (omit `model` field).
+- Built-in agents: `default`, `worker`, `explorer`. Custom agents extend these via TOML definitions.
 - Codex operates under network restrictions — skills should not assume outbound HTTP access.
 - Use `$ARGUMENTS` to access user-provided arguments when the skill is invoked.
 - All skill guidance executes within the sandbox; file I/O is confined to the workspace.
+- Skills are installed at `.agents/skills/<skill-id>/SKILL.md`. Workflow skills can also be compiled to `.agents/skills/<workflow-id>/SKILL.md` as `generatedSkills`.
+- Codex supports three autonomy levels: `suggest`, `auto-edit`, `full-auto`.
+- MCP skill tools are available when the Cubis Foundry MCP server is connected.