oh-my-opencode 5.0.0-beta.8 → 5.0.0-beta.80

package/packages/omo-codex/plugin/skills/frontend/references/design/print-paged-media.md

@@ -0,0 +1,76 @@
+---
+name: print-paged-media
+description: "Layer A paged-media reference. Load it when output is a document that breaks into pages - a PDF report, a print stylesheet, a headless-Chrome print pipeline, or an HTML deck printed to paper - instead of a continuously scrolling screen. Owns the page box, fragmentation properties, the atomic-block set, and the keep-together side effect that strands a block on a near-empty page. Adds zero visual taste; the style skill and DESIGN.md still own color, type, and spacing."
+---
+
+# Paged Media
+
+Screen layout has one viewport and infinite scroll. Paged media has a fixed page box and a fragmentation engine that cuts your content into pages whether or not you told it where. This file owns that cut. It stacks on top of a Layer A style skill and adds ZERO visual direction - color, type, radius, and spacing still come from the style skill and `DESIGN.md`.
+
+Load this when the deliverable is a PDF report, a print stylesheet, a `chrome --headless --print-to-pdf` pipeline, a WeasyPrint render, or any HTML meant to land on paper.
+
+## 1. The page box comes first
+
+Declare the page before styling anything inside it. `@page { size: A4; margin: 2cm; }` sets the sheet and the content area; every measurement downstream is relative to what survives those margins.
+
+- `@page:first` targets the cover alone. A full-bleed cover is `@page:first { margin: 0 }`, which also suppresses the running header and footer on page one.
+- Running headers and footers live in the page margin boxes. In headless Chrome they come from `headerTemplate` / `footerTemplate` on `Page.printToPDF`, rendered into the margins that `@page` declared - not from a fixed-position element in the body.
+- Pass `preferCSSPageSize: true` when printing through CDP so your `@page` rule wins over the caller's paper defaults. Without it, the size you declared and the size you get can differ.
+- Body content flows through the page area. An element with a viewport unit (`100vh`) or a fixed position does not mean on paper what it means on screen; size page furniture in `mm` or `pt`.
+
+## 2. Fragmentation vocabulary
+
+Three properties decide every cut. Modern names first; the `page-break-*` aliases still work and appear in older pipelines.
+
+| Property | Use it for |
+|---|---|
+| `break-before: page` | Start a section on a fresh page. |
+| `break-after: avoid` | Keep a heading with the block that follows it (keep-with-next). |
+| `break-inside: avoid` | Forbid a cut *through* one block: a table, a figure, a callout, a code listing. |
+| `orphans` / `widows` | Minimum lines of a paragraph left at a page foot / carried to the next page. Set both to at least 2; 3 reads better in dense body copy. |
+
+`break-inside: avoid` on a wrapper is what actually protects a table - apply it to the element that owns the whole exhibit, not to the `<table>` alone, or the caption and the rules can still separate from the rows.
+
+## 3. The atomic set - keep it small on purpose
+
+Mark as atomic only what is unreadable when split: a data table with its caption, a figure with its `figcaption`, a callout, a short code listing. Everything else - ordinary paragraphs, long prose lists, multi-page tables - should be allowed to break, because a block that cannot break and cannot fit is a defect generator, not a protected exhibit.
+
+A block taller than the page area gets split regardless of what you declared. `avoid` is a request the engine honors only when honoring it is possible.
+
+## 4. The keep-together side effect - the defect this file exists to prevent
+
+This is the failure that looks like a bug in your CSS but is your CSS working exactly as written.
+
+A block carrying `break-inside: avoid` reaches the bottom of a page with too little room left. The engine cannot split it, so it moves the WHOLE block to the next page. That is correct. The damage appears when the block is the LAST content of its section and the next section carries `break-before: page`: nothing can flow up to backfill the gap it left, and nothing can flow down beside it. You ship one page with a short box at the top and a column of whitespace under it, and one page before it with an unexplained gap at the foot.
+
+Nothing in the cascade detects this. Both rules are locally correct; the artifact is their interaction with where the text happened to land.
+
+How to prevent and repair it:
+
+- **Shrink the atomic set.** Every unnecessary `break-inside: avoid` is one more block that can be stranded. Protect exhibits, not paragraphs.
+- **Do not wrap a lead-in with its block** unless the pair reliably fits together. Binding more content into one unbreakable unit makes stranding more likely, not less.
+- **Reconsider the forced section break.** `break-before: page` on every section is a strong default for a report and a poor one for short trailing sections; allowing a brief section to continue on the current page removes the trap entirely.
+- **Editorial repair is legitimate.** Shortening the block by a line, or moving it a paragraph earlier, is a real fix - and often the correct one for a hand-tuned deliverable.
+- **Detect it by ink, not by intent.** A page whose content coverage falls far below its neighbors is the signal. Ranking pages by rendered coverage finds stranded blocks faster than reading CSS does.
+
+## 5. What the engine actually honors
+
+Fragmentation support is uneven, and reading the stylesheet will not tell you what happened.
+
+- `break-inside: avoid` inside flex, grid, and multi-column containers is partially implemented across engines; a constraint that holds in normal flow can be dropped inside a nested fragmentation context.
+- Transformed, absolutely positioned, and `overflow`-clipped subtrees fragment differently from ordinary blocks, and sometimes not at all.
+- Background colors and images on split blocks may or may not repaint on the continuation page. Chrome needs `printBackground: true` before any of it renders.
+
+Treat every fragmentation rule as a hypothesis until you have seen the rendered page.
+
+## 6. Verify by rendering every page
+
+A paged deliverable is verified by looking at its pages. Render them and inspect each one:
+
+```bash
+pdftoppm -png -r 150 report.pdf out/page   # one PNG per page
+```
+
+Read every page image for: blank or near-empty pages, a block stranded alone after a keep-together push, a table or figure split mid-exhibit, a heading orphaned at a page foot, clipped or overflowing text, and a running header or footer that drifted or vanished.
+
+Extracting the text proves the words are present. It cannot see any defect in this list - layout is exactly the information that extraction discards. Fix, re-render, and look again until the pages are clean; the same standard governs the `ulw-research` delivery gate and the `mass-ulw` verification wave.

package/packages/omo-codex/plugin/skills/frontend/references/designpowers/README.md

@@ -9,7 +9,7 @@ The purpose of this ruleset is to enrich the existing frontend workflow while pr
 - `references/design/README.md` still owns the `DESIGN.md` contract, taste routing, brand references, React dev tooling, and browser-based design QA expectations.
 - `references/perfection/README.md` still owns Lighthouse, performance, SEO, accessibility audit mechanics, and real-browser verification.
 - `/visual-qa` still owns objective rendered evidence for visual claims.
-- `/ulw-plan`, `/start-work`, and `/review-work` still own planning, execution, and final implementation review when those phases apply.
+- `/ulw-plan`, `/ulw-execute`, and `/review-work` still own planning, execution, and final implementation review when those phases apply.
 
 ## Load Order
 

package/packages/omo-codex/plugin/skills/frontend/references/designpowers/lane-a-direction.md

@@ -61,4 +61,4 @@ Lane A passes only when the plan has inspectable design context, not vague inten
 
 PASS when `ulw-plan` receives this lane context and produces a decision-complete plan whose tasks are design-aware, persona-aware, and evidence-bound.
 
-FAIL when the plan skips inclusive-personas, buries accessibility as a final polish item, treats taste as generic style, writes a parallel design plan, or leaves design decisions for start-work workers to invent.
+FAIL when the plan skips inclusive-personas, buries accessibility as a final polish item, treats taste as generic style, writes a parallel design plan, or leaves design decisions for ulw-execute workers to invent.

package/packages/omo-codex/plugin/skills/frontend/references/designpowers/lane-b-execution.md

@@ -1,25 +1,25 @@
 # Lane B: Design Execution Guidance
 
-Lane B feeds `start-work` workers and always loads `frontend` for UI implementation. It does not install a separate builder. Its job is to carry designpowers composition, interaction, motion, content, responsive, adaptive, token, and cognitive-accessibility guidance into worker prompts while keeps ownership of decomposition, implementation, QA, evidence, and ledger updates.
+Lane B feeds `ulw-execute` workers and always loads `frontend` for UI implementation. It does not install a separate builder. Its job is to carry designpowers composition, interaction, motion, content, responsive, adaptive, token, and cognitive-accessibility guidance into worker prompts while keeps ownership of decomposition, implementation, QA, evidence, and ledger updates.
 
 ## Phase Owner
 
 | Capability | Materialized designpowers source | owner | Mapping |
 |---|---|---|---|
-| Visual hierarchy, layout, color, typography, touch targets, and WCAG contrast | `ui-composition` | `start-work` worker plus `frontend` | Add visual constraints and acceptance checks to worker assignments. `frontend` remains the required UI implementation skill. |
-| States, feedback, loading, error, keyboard, touch, and recovery behavior | `interaction-design` | `start-work` worker plus `frontend` | Require every interactive component to define default, hover, focus, active, disabled, loading, success, and error states where applicable. |
-| Purposeful motion and reduced-motion alternatives | `motion-choreography` | `start-work` worker plus `frontend` | Require motion specs to explain what changed, what to look at next, or how elements relate; require safe alternatives. |
-| Content-driven breakpoints and zoom behavior | `responsive-patterns` | `start-work` worker plus `frontend` | Require responsive proof at narrow, mid, desktop, and 200 percent zoom scenarios when the surface is visual. |
-| User preference adaptation | `adaptive-interfaces` | `start-work` worker plus `frontend` | Require support for relevant preferences such as color scheme, contrast, reduced motion, text sizing, and density. |
-| Mental load, wayfinding, focus management, memory demands, and recovery paths | `cognitive-accessibility` | `start-work` worker plus `frontend` | Add COGA-style checks directly to worker acceptance criteria for flows, forms, navigation, and dense tools. |
-| Plain-language labels, headings, alt text, link text, errors, and instructions | `accessible-content` and `voice-and-tone` | `start-work` worker plus `frontend` | Require final copy, no placeholders, useful errors, readable labels, and consistent tone. |
+| Visual hierarchy, layout, color, typography, touch targets, and WCAG contrast | `ui-composition` | `ulw-execute` worker plus `frontend` | Add visual constraints and acceptance checks to worker assignments. `frontend` remains the required UI implementation skill. |
+| States, feedback, loading, error, keyboard, touch, and recovery behavior | `interaction-design` | `ulw-execute` worker plus `frontend` | Require every interactive component to define default, hover, focus, active, disabled, loading, success, and error states where applicable. |
+| Purposeful motion and reduced-motion alternatives | `motion-choreography` | `ulw-execute` worker plus `frontend` | Require motion specs to explain what changed, what to look at next, or how elements relate; require safe alternatives. |
+| Content-driven breakpoints and zoom behavior | `responsive-patterns` | `ulw-execute` worker plus `frontend` | Require responsive proof at narrow, mid, desktop, and 200 percent zoom scenarios when the surface is visual. |
+| User preference adaptation | `adaptive-interfaces` | `ulw-execute` worker plus `frontend` | Require support for relevant preferences such as color scheme, contrast, reduced motion, text sizing, and density. |
+| Mental load, wayfinding, focus management, memory demands, and recovery paths | `cognitive-accessibility` | `ulw-execute` worker plus `frontend` | Add COGA-style checks directly to worker acceptance criteria for flows, forms, navigation, and dense tools. |
+| Plain-language labels, headings, alt text, link text, errors, and instructions | `accessible-content` and `voice-and-tone` | `ulw-execute` worker plus `frontend` | Require final copy, no placeholders, useful errors, readable labels, and consistent tone. |
 | Tokens and design-system consistency | `token-architecture` and `design-system-alignment` | `frontend` | Require real design tokens and existing component patterns before new one-off styling. |
 
 Materialized agent references for this lane: `design-lead`, `motion-designer`, and `content-writer`. They are prompt-role references for spawned OpenAgent work, not alternate executors.
 
 ## Prompt Injection
 
-Add this block to each `start-work` implementation worker that touches UI:
+Add this block to each `ulw-execute` implementation worker that touches UI:
 
 ```text
 Load `frontend` for UI implementation. Also apply Lane B Design Execution Guidance.
@@ -43,7 +43,7 @@ Do not invent visual direction that conflicts with the plan. If the plan lacks a
 Lane B worker DoneClaims must include:
 
 - Exact changed files and the `frontend` references loaded.
-- The real-surface QA invocation required by `start-work` for the UI surface, with captured artifact path.
+- The real-surface QA invocation required by `ulw-execute` for the UI surface, with captured artifact path.
 - Screenshot, browser, HTTP, or xterm.js web-terminal artifacts appropriate to the visible surface.
 - Accessibility evidence from the existing OpenAgent frontend path, such as Lighthouse, react-doctor, keyboard checks, or other plan-required checks.
 - A short design trace: which persona, design principle, token, or state requirement each major UI decision satisfies.
@@ -52,7 +52,7 @@ Lane B worker DoneClaims must include:
 ## Guardrails
 
 - UI implementation always goes through `frontend`; Lane B only enriches the worker prompt.
-- `start-work` owns decomposition, worker dispatch, evidence ledger entries, adversarial checks, and completion state.
+- `ulw-execute` owns decomposition, worker dispatch, evidence ledger entries, adversarial checks, and completion state.
 - The materialized `design-builder` concept is not used as an available executor. OpenAgent workers build.
 - No placeholders, generic copy, unverified contrast claims, decorative-only motion, or one-off hardcoded design systems pass this lane.
 - Accessibility and cognitive-accessibility are implementation constraints, not review-only cleanup.

package/packages/omo-codex/plugin/skills/frontend/references/designpowers/lane-d-memory.md

@@ -7,7 +7,7 @@ Lane D records the design record around OpenAgent work. It owns `.omo/frontend-d
 | Capability | Source boundary | owner | Mapping |
 |---|---|---|---|
 | Track deferred design and accessibility findings | `design-debt-tracker` | `.omo/frontend-design/state.md` plus final `review-work` context | Maintain a register with ID, date, source, severity, issue, affected users, suggested fix, status, and notes. |
-| Package design rationale for implementers or reviewers | `design-handoff` | `start-work` and `review-work` context | Record component, interaction, accessibility, content, and rationale notes that workers or reviewers need. |
+| Package design rationale for implementers or reviewers | `design-handoff` | `ulw-execute` and `review-work` context | Record component, interaction, accessibility, content, and rationale notes that workers or reviewers need. |
 | Reflect on process after completion | `design-retrospective` | final handoff context | Record what worked, what did not, fix rounds, debt health, and lessons. |
 | Maintain observational design memory | Authored concept preserved from excluded raw `design-memory` | `.omo/frontend-design/state.md` and optional personal report context | Store observations descriptively. Do not feed them back as future project constraints. |
 | Produce a taste reflection when requested | `taste-report` | user-facing handoff only | Summarize personal-layer observations only when enough evidence exists or the user asks. |
@@ -65,7 +65,7 @@ Lane D passes only when the record is inspectable:
 - Accessibility debt is either resolved or explicitly acknowledged by the user before acceptance.
 - Handoff notes cite concrete artifacts, decisions, constraints, and evidence paths.
 - Retrospective notes cite the final verification artifacts, unresolved debt, and lessons from fix rounds or user overrides.
-- The evidence index points to real files produced by OpenAgent planning, `start-work`, `visual-qa`, or `review-work`.
+- The evidence index points to real files produced by OpenAgent planning, `ulw-execute`, `visual-qa`, or `review-work`.
 
 ## Guardrails
 
@@ -78,6 +78,6 @@ Lane D passes only when the record is inspectable:
 
 ## Pass / Fail Behavior
 
-PASS when state, design debt, handoff, retrospective, and evidence references are current enough for `ulw-plan`, `start-work`, `visual-qa`, or `review-work` to resume without guessing.
+PASS when state, design debt, handoff, retrospective, and evidence references are current enough for `ulw-plan`, `ulw-execute`, `visual-qa`, or `review-work` to resume without guessing.
 
 FAIL when deferred findings disappear, accessibility debt is accepted without acknowledgement, handoff omits artifact paths, state is stale, retrospective claims lack evidence, or memory is used as prescriptive design input.

package/packages/omo-codex/plugin/skills/frontend/references/designpowers/orchestration.md

@@ -51,7 +51,7 @@ When using a designpowers role reference:
 
 - name the role in the prompt text;
 - include a self-contained task, deliverable, scope, and verification expectation;
-- route actual phase ownership to frontend design/perfection, `/ulw-plan`, `/start-work`, `/visual-qa`, or `/review-work`;
+- route actual phase ownership to frontend design/perfection, `/ulw-plan`, `/ulw-execute`, `/visual-qa`, or `/review-work`;
 - record findings in the state file or review packet only when backed by artifacts or located observations.
 
 ## Reconciliation Ladder

package/packages/omo-codex/plugin/skills/frontend/references/designpowers/routing.md

@@ -1,13 +1,13 @@
 # designpowers Routing Contract
 
-`designpowers` routes design operating-layer guidance inside the existing frontend skill. It is not a replacement for `/frontend`, `/visual-qa`, `/ulw-plan`, `/start-work`, `/review-work`, or any OpenAgent skill, and it must not create a second planner, builder, verification harness, or orchestration API.
+`designpowers` routes design operating-layer guidance inside the existing frontend skill. It is not a replacement for `/frontend`, `/visual-qa`, `/ulw-plan`, `/ulw-execute`, `/review-work`, or any OpenAgent skill, and it must not create a second planner, builder, verification harness, or orchestration API.
 
 ## Phase Routing
 
 | User intent or workflow phase | Load or instruct | Required handoff from designpowers |
 |---|---|---|
 | Ambiguous or multi-step web UI request; any request needing a plan | `/ulw-plan` plus this reference's `lane-a-direction.md` | Provide design discovery prompts, target users, inclusive personas, taste direction, open owner decisions, and design debt policy as planning inputs. |
-| Approved plan execution; continuing an OpenAgent plan | `/start-work` plus `lane-b-execution.md` | Keep execution under Boulder/ledger discipline and include current design-state constraints in worker assignments. |
+| Approved plan execution; continuing an OpenAgent plan | `/ulw-execute` plus `lane-b-execution.md` | Keep execution under Boulder/ledger discipline and include current design-state constraints in worker assignments. |
 | Building, styling, redesigning, auditing, or performance-checking a web UI | frontend `design` + `perfection`; add `lane-b-execution.md` when designpowers context affects implementation | Preserve the frontend `DESIGN.md` gate, design/perfection routing, React tooling, real-browser checks, and implementation standards. |
 | Screenshots, visual regressions, clone fidelity, layout quality, alpha/CJK checks, or design QA | `/visual-qa` plus `lane-c-review.md` | Run objective evidence capture before design judgment and feed the same artifacts into persona/accessibility/heuristic review. |
 | Final implementation approval, QA my work, review changes, or significant completed implementation | `/review-work` plus `lane-c-review.md` and `lane-d-memory.md` | Include the design brief, state file path, visual artifacts, unresolved design debt, and accessibility-debt acknowledgements as review inputs. |
@@ -25,9 +25,9 @@ When planning is needed, `designpowers` supplies design-specific context and let
 - verification expectations: frontend checks, visual QA artifacts, persona walkthroughs, and review-work sign-off;
 - explicit Must Not Have constraints, including prohibited bridge/canvas tooling.
 
-## Execution Through `/start-work`
+## Execution Through `/ulw-execute`
 
-When a plan is approved or selected, `/start-work` remains the orchestrator. `designpowers` only enriches worker prompts with design context from `.omo/frontend-design/state.md` and the selected plan. Worker prompts should carry:
+When a plan is approved or selected, `/ulw-execute` remains the orchestrator. `designpowers` only enriches worker prompts with design context from `.omo/frontend-design/state.md` and the selected plan. Worker prompts should carry:
 
 - the exact plan checkbox and files in scope;
 - design-state constraints that affect the task;
@@ -35,7 +35,7 @@ When a plan is approved or selected, `/start-work` remains the orchestrator. `de
 - required `/visual-qa` loading for rendered visual proof;
 - the design debt rule: unresolved accessibility debt cannot disappear into a summary.
 
-Direct implementation outside `/start-work` is not part of this routing contract when a Prometheus plan is active.
+Direct implementation outside `/ulw-execute` is not part of this routing contract when a Prometheus plan is active.
 
 ## UI Build Through Frontend `design` And `perfection`
 

package/packages/omo-codex/plugin/skills/frontend/scripts/perfection/lighthouse-audit.py

@@ -11,8 +11,8 @@
 # ─── How to run ───
 # 1. Install uv (if not installed):
 #      curl -LsSf https://astral.sh/uv/install.sh | sh
-# 2. Install Playwright browsers (one-time):
-#      uv run --with playwright python -m playwright install chromium
+# 2. Chrome must already be installed locally:
+#      The script uses channel="chrome"; it downloads no browser.
 # 3. Run:
 #      uv run lighthouse-audit.py https://example.com
 #      uv run lighthouse-audit.py https://example.com --desktop-only

package/packages/omo-codex/plugin/skills/git-master/SKILL.md

@@ -1,6 +1,6 @@
 ---
 name: git-master
-description: "MUST USE whenever a task needs a commit or git-history investigation. Covers atomic commits, staging, commit-message style, rebase, squash, fixup/autosquash, blame, bisect, reflog, git log -S/-G, and questions like who wrote this or when was this added. Do not use for ordinary code edits unless the user asks for git work."
+description: "Handles git work: atomic commits, rebase, squash, blame, bisect, reflog, and history questions. Use whenever a task needs a commit or a git-history investigation; skip for ordinary code edits."
 ---
 
 # Git Master

package/packages/omo-codex/plugin/skills/init-deep/SKILL.md

@@ -1,6 +1,6 @@
 ---
 name: init-deep
-description: "(builtin) Initialize hierarchical AGENTS.md knowledge base"
+description: "Initializes a hierarchical AGENTS.md knowledge base for a project. Use when a repo needs its structure, commands, and conventions documented for agents."
 ---
 ## Codex Harness Tool Compatibility
 
@@ -11,17 +11,19 @@ This skill may include examples copied from the OpenCode harness. In Codex, do n
 | `call_omo_agent(subagent_type="explore", ...)` | `multi_agent_v1.spawn_agent({"message":"TASK: act as an explorer. ...","agent_type":"explorer","fork_context":false})` |
 | `call_omo_agent(subagent_type="librarian", ...)` | `multi_agent_v1.spawn_agent({"message":"TASK: act as a librarian. ...","agent_type":"librarian","fork_context":false})` |
 | `task(subagent_type="plan", ...)` | `multi_agent_v1.spawn_agent({"message":"TASK: act as a planning agent. ...","agent_type":"plan","fork_context":false})` |
-| `task(subagent_type="oracle", ...)` for final verification | `multi_agent_v1.spawn_agent({"message":"TASK: act as a rigorous reviewer. ...","agent_type":"lazycodex-gate-reviewer","fork_context":false})` |
-| `task(category="...", ...)` for implementation or QA | `multi_agent_v1.spawn_agent({"message":"TASK: act as an implementation or QA worker. ...","fork_context":false})` |
+| `task(subagent_type="oracle", ...)` for final verification | By default, record a self-review in the notepad: re-read the diff, run diagnostics, and capture evidence for every acceptance criterion. Only when the user demanded strict, rigorous, or high-accuracy review, use `multi_agent_v1.spawn_agent({"message":"TASK: act as a rigorous reviewer. ...","agent_type":"lazycodex-gate-reviewer","fork_context":false})`; include `fork_context: false`. |
+| `task(category="...", ...)` for implementation or QA | `multi_agent_v1.spawn_agent({"message":"TASK: act as an implementation or QA worker. ...","agent_type":"lazycodex-worker-medium","fork_context":false})` |
 | `background_output(task_id="...")` | `multi_agent_v1.wait_agent(...)` for mailbox signals |
 | `team_*(...)` | Use Codex native subagents via `multi_agent_v1.spawn_agent` and `multi_agent_v1.wait_agent`; use `multi_agent_v1.send_input` and `multi_agent_v1.close_agent` only when exposed in the active tools list |
 
-Role-specific behavior must be described in a self-contained `message`. Use `fork_context: false` to start the child with only the initial prompt (no parent history); use `fork_context: true` only when full parent history is truly required. Include any required conversation context, files, diffs, constraints, and requested skill names directly in the spawned agent's `message`. OMO installs these selectable agent roles into `~/.codex/agents/`: `explorer`, `librarian`, `plan`, `momus`, `metis`, `lazycodex-code-reviewer`, `lazycodex-qa-executor`, and `lazycodex-gate-reviewer` - pass the matching name as `agent_type` so the child gets that role's model and instructions. If the spawn tool exposes no `agent_type` parameter, omit it and describe the role inside `message`. If a code block below conflicts with this section, this section wins.
+Role-specific behavior must be described in a self-contained `message`. Use `fork_context: false` to start the child with only the initial prompt (no parent history); use `fork_context: true` only when full parent history is truly required. Include any required conversation context, files, diffs, constraints, and requested skill names directly in the spawned agent's `message`. OMO installs these selectable agent roles into `~/.codex/agents/`: `explorer`, `librarian`, `plan`, `momus`, `metis`, `lazycodex-code-reviewer`, `lazycodex-qa-executor`, and `lazycodex-gate-reviewer` - pass the matching name as `agent_type` so the child gets that role's model and instructions. Inspect the actual spawn schema: whenever `agent_type` is exposed, EVERY spawn MUST select an exact LazyCodex role, on V1 or V2. Implementation difficulty selects `lazycodex-worker-low`, `lazycodex-worker-medium`, or `lazycodex-worker-high`; clone QA can select `lazycodex-clone-fidelity-reviewer`. Never select generic `worker` or `default`. If a code block below conflicts with this section, this section wins.
 
-Codex exposes ONE of two subagent tool surfaces per session; check your own tool list and route accordingly. If `multi_agent_v1.*` tools exist, use the table above as written. If instead a flat `spawn_agent` with a required `task_name` exists (`multi_agent_v2`), rewrite every `multi_agent_v1.*` example: `multi_agent_v1.spawn_agent({...,"fork_context":false})` becomes `spawn_agent({"task_name":"<lowercase_digits_underscores>","message":...,"agent_type":...,"fork_turns":"none"})` (`"all"` only when full parent history is truly required); `send_input` becomes `send_message`; do not call `close_agent`/`resume_agent` (finished agents end on their own; `followup_task` re-tasks one, `interrupt_agent` stops one); `wait_agent` takes only `timeout_ms` and returns on any child mailbox activity. On the v2 surface `agent_type` may be ABSENT from the spawn schema (verified 2026-07-11: only `fork_turns`/`message`/`task_name`) — when absent, omit it and describe the role inside `message`; installed role TOMLs cannot be selected on that surface. If a code block below conflicts with this section, this section wins.
+Codex exposes ONE of two subagent tool surfaces per session; check your own tool list and route accordingly. If `multi_agent_v1.*` tools exist, use the table above as written. If instead a flat `spawn_agent` with a required `task_name` exists (`multi_agent_v2`), rewrite every `multi_agent_v1.*` example: `multi_agent_v1.spawn_agent({...,"fork_context":false})` becomes `spawn_agent({"task_name":"<lowercase_digits_underscores>","message":...,"agent_type":...,"fork_turns":"none"})` (`"all"` only when full parent history is truly required); `send_input` becomes `send_message`; do not call `close_agent`/`resume_agent` (finished agents end on their own; `followup_task` re-tasks one, `interrupt_agent` stops one); `wait_agent` takes only `timeout_ms` and returns on any child mailbox activity. Do not infer role support from V1/V2 or a model version. Legacy-schema exception: only if the actual schema lacks `agent_type`, omit that unsupported field and carry the complete role instructions in `message`, with history explicitly disabled. This cannot select a specialized TOML; an installed managed default supplies the medium worker for unnamed non-forks. The guard has no schema metadata and rejects unnamed calls, so report incompatible routing instead of retrying generically. Every deliberate full-history fork must still name its role: Codex skips role application on unnamed full-history forks, an upstream gap no LazyCodex default can repair. If a code block below conflicts with this section, this section wins. `fork_context` is rejected on `multi_agent_v2` (`fork_context is not supported in MultiAgentV2; use fork_turns instead`).
 
 When translating `load_skills=[...]`, include the requested skill names in the spawned agent's `message`. If a code block below conflicts with this section, this section wins.
 
+Omit optional keys you do not set. Never send `items: []`, `message: ""`, `model: ""`, `reasoning_effort: ""`, or `service_tier: ""` — Codex rejects them (`Items can't be empty`, `reasoning_effort must not be empty`).
+
 For work likely to exceed one wait cycle, require the child to send `WORKING: <task> - <current phase>` before long passes and `BLOCKED: <reason>` only when progress stops. A `multi_agent_v1.wait_agent` timeout only means no new mailbox update arrived; back off between waits (double the timeout up to ~5 minutes) instead of spinning short cycles. Treat a running child as alive. Fallback only when the child is completed without the deliverable, ack-only after followup, explicitly `BLOCKED:`, or no longer running.
 
 # /init-deep
@@ -42,7 +44,7 @@ Generate hierarchical AGENTS.md files. Root + complexity-scored subdirectories.
 
 1. **Discovery + Analysis** (concurrent)
    - Fire background explore agents immediately
-   - Main session: bash structure + LSP/codegraph code map + read existing AGENTS.md
+   - Main session: bash structure + LSP/ast-grep code map + read existing AGENTS.md
 2. **Score & Decide** - Determine AGENTS.md locations from merged findings
 3. **Generate** - Root first, then subdirs in parallel
 4. **Review** - Deduplicate, trim, validate
@@ -51,7 +53,7 @@ Generate hierarchical AGENTS.md files. Root + complexity-scored subdirectories.
 **TodoWrite ALL phases. Mark in_progress → completed in real-time.**
 ```
 TodoWrite([
-  { id: "discovery", content: "Fire explore agents + LSP/codegraph map + read existing", status: "pending", priority: "high" },
+  { id: "discovery", content: "Fire explore agents + LSP/ast-grep map + read existing", status: "pending", priority: "high" },
   { id: "scoring", content: "Score directories, determine locations", status: "pending", priority: "high" },
   { id: "generate", content: "Generate AGENTS.md files (root + subdirs)", status: "pending", priority: "high" },
   { id: "review", content: "Deduplicate, validate, trim", status: "pending", priority: "medium" }
@@ -67,16 +69,16 @@ TodoWrite([
 
 ### Fire Background Explore Agents IMMEDIATELY
 
-Don't wait-these run async while main session works. **Equip every agent with the code graph**: any task touching structure, entry points, dependencies, or hotspots MUST query `codegraph_*` (explore/search/callers/callees/impact) and `lsp_symbols` when present, and ground its claims in that data instead of guessing from conventions. Richer real-graph context per agent = a more accurate project map.
+Don't wait-these run async while main session works. **Ground every agent in real symbol data**: any task touching structure, entry points, dependencies, or hotspots MUST query `lsp_symbols`/`lsp_find_references` and the ast-grep skill (`sg`) when present, and ground its claims in that data instead of guessing from conventions. Richer real context per agent = a more accurate project map.
 
 ```
 // Fire all at once, collect results later
-task(subagent_type="explore", load_skills=[], description="Explore project structure", run_in_background=true, prompt="Project structure: map real layout via codegraph_explore/codegraph_files → REPORT deviations from standard patterns")
-task(subagent_type="explore", load_skills=[], description="Find entry points", run_in_background=true, prompt="Entry points: FIND main files, trace reach via codegraph_callees + lsp_symbols → REPORT non-standard organization")
+task(subagent_type="explore", load_skills=[], description="Explore project structure", run_in_background=true, prompt="Project structure: map real layout via glob + lsp_symbols scope=workspace → REPORT deviations from standard patterns")
+task(subagent_type="explore", load_skills=[], description="Find entry points", run_in_background=true, prompt="Entry points: FIND main files, trace reach via lsp_symbols + lsp_find_references → REPORT non-standard organization")
 task(subagent_type="explore", load_skills=[], description="Find conventions", run_in_background=true, prompt="Conventions: FIND config files (.eslintrc, pyproject.toml, .editorconfig) → REPORT project-specific rules")
 task(subagent_type="explore", load_skills=[], description="Find anti-patterns", run_in_background=true, prompt="Anti-patterns: FIND 'DO NOT', 'NEVER', 'ALWAYS', 'DEPRECATED' comments → LIST forbidden patterns")
 task(subagent_type="explore", load_skills=[], description="Explore build/CI", run_in_background=true, prompt="Build/CI: FIND .github/workflows, Makefile → REPORT non-standard patterns")
-task(subagent_type="explore", load_skills=[], description="Find test patterns", run_in_background=true, prompt="Test patterns: FIND test configs/structure; codegraph_callers on core modules to see what is covered → REPORT unique conventions")
+task(subagent_type="explore", load_skills=[], description="Find test patterns", run_in_background=true, prompt="Test patterns: FIND test configs/structure; lsp_find_references on core modules to see what is covered → REPORT unique conventions")
 ```
 
 <dynamic-agents>
@@ -138,7 +140,7 @@ For each existing file found:
 
 If `--create-new`: Read all existing first (preserve context) → then delete all → regenerate.
 
-#### 3. Code Map - drive LSP AND codegraph (do NOT skip)
+#### 3. Code Map - drive LSP AND ast-grep (do NOT skip)
 
 Highest-signal source for the CODE MAP and the Symbol/Export/Reference scoring rows. Complementary, not alternatives - run BOTH when present, alongside the explore agents.
 
@@ -147,10 +149,10 @@ Highest-signal source for the CODE MAP and the Symbol/Export/Reference scoring r
 - `lsp_symbols` scope="workspace", query by kind (class/interface/function) -> symbol inventory.
 - `lsp_find_references` on top exports (line/character from the symbols result) -> reference centrality.
 
-**codegraph** - when `codegraph_*` tools exist (check `codegraph_status`); a first-class peer to LSP, NOT a last resort:
-- `codegraph_explore` -> overview; `codegraph_callers`/`codegraph_callees`/`codegraph_impact` -> centrality + blast radius for the scoring matrix; `codegraph_search`/`codegraph_files` -> symbol/file inventory.
+**ast-grep** - load the ast-grep skill; a first-class peer to LSP, NOT a last resort:
+- `sg --pattern 'export $$$' --lang <lang>` -> export inventory; `sg --pattern 'import $$$ from "$MOD"'` -> dependency edges between modules for the scoring matrix.
 
-Only if NEITHER exists: explore agents + the ast-grep skill (`sg`), and mark centrality unmeasured in the CODE MAP.
+Only if LSP is unavailable: explore agents + ast-grep alone, and mark reference centrality unmeasured in the CODE MAP.
 
 ### Collect Background Results
 
@@ -159,7 +161,7 @@ Only if NEITHER exists: explore agents + the ast-grep skill (`sg`), and mark cen
 for each background task ID (`bg_...`): background_output(task_id="bg_...")
 ```
 
-**Merge: bash + LSP/codegraph + existing + explore findings. Mark "discovery" as completed.**
+**Merge: bash + LSP/ast-grep + existing + explore findings. Mark "discovery" as completed.**
 
 ---
 
@@ -235,7 +237,7 @@ NEVER use Write to overwrite an existing file. ALWAYS check existence first via
 |------|----------|-------|
 
 ## CODE MAP
-{From LSP/codegraph - skip only if neither exists or project <10 files}
+{From LSP/ast-grep - skip only if neither exists or project <10 files}
 
 | Symbol | Type | Location | Refs | Role |
 |--------|------|----------|------|------|
@@ -318,7 +320,7 @@ Hierarchy:
 ## Anti-Patterns
 
 - **Static agent count**: MUST vary agents based on project size/depth
-- **Sequential execution**: MUST parallel (explore + LSP + codegraph concurrent)
+- **Sequential execution**: MUST parallel (explore + LSP + ast-grep concurrent)
 - **Ignoring existing**: ALWAYS read existing first, even with --create-new
 - **Over-documenting**: Not every dir needs AGENTS.md
 - **Redundancy**: Child never repeats parent

package/packages/omo-codex/plugin/skills/lcx-doctor/SKILL.md

@@ -66,11 +66,12 @@ sync_latest_source openai/codex "$LAZYCODEX_SOURCE_ROOT/openai-codex-source"
    - Installed LazyCodex version: the `version` in the installed plugin manifest, discoverable with `find "${CODEX_HOME:-$HOME/.codex}/plugins" -path '*/.codex-plugin/plugin.json'`. Installed plugins live under `$CODEX_HOME/plugins/cache/<marketplace>/<name>/<version>/`.
    - Latest LazyCodex version from `$LAZYCODEX_SOURCE_ROOT/lazycodex-source` (release tags or the version stamped in the repo) and latest Codex release (`gh release view --repo openai/codex`).
    - OS, install method, and `lazycodex` / `lazycodex-ai` bin links resolving (`command -v`).
+   - Astra readiness. The LazyCodex catalog default is `gpt-6-astra`, which Codex only knows from codex-cli 0.153.1 onward. First release tag containing the `models.json` change: `rust-v0.153.1` (backport commit `5cc1c94b8e`). Reproduce on a full clone with tags: `git -C <codex-clone> log --format='%h %d' -S gpt-6-astra -- codex-rs/models-manager/models.json`, then `git -C <codex-clone> tag --contains 5cc1c94b8e | sort -V | head -1`; a shallow clone or one without tags returns nothing. Record the installed version from `codex --version`, the root `model` in `$CODEX_HOME/config.toml` (informational only, it doesn't change the verdict), and, when `$CODEX_HOME/models_cache.json` exists, whether it contains `gpt-6-astra` (`grep -c gpt-6-astra "${CODEX_HOME:-$HOME/.codex}/models_cache.json"`). Verdict: PASS when `codex --version` is 0.153.1 or newer and `models_cache.json` is either absent or lists `gpt-6-astra`. WARN when the version is older than 0.153.1, or the cache exists but lacks `gpt-6-astra`; remediation is upgrading Codex (`npm install -g @openai/codex@latest`, or the package manager that installed `codex`), then starting Codex once so `models_cache.json` refreshes.
 3. Check config and wiring against the latest installer, not against assumptions. Read what the current installer under `$LAZYCODEX_SOURCE_ROOT/lazycodex-source` writes (installer sources live in the omo-codex package, e.g. `scripts/install/`), then verify the local equivalents:
    - `$CODEX_HOME/config.toml` exists and parses; LazyCodex-managed entries match what the latest installer would write.
    - Plugin payload present and non-empty: read `.codex-plugin/plugin.json`; when that manifest declares a `hooks` array, validate every direct hook path declared by the manifest; require `hooks/hooks.json` only when the manifest declares it; do not require retired paths such as `components/workflow-selector` or `hooks/user-prompt-submit-selecting-lazycodex-workflow.json` unless the current manifest declares them.
    - Verify the manifest-declared runtime payload, not a remembered source tree. Current payload includes `skills/`, `.mcp.json`, root CLI runtimes such as `dist/cli/index.js` and `dist/cli-node/index.js`, and every hook/MCP `components/*/dist/*.js` target referenced by installed manifests.
-   - Treat install-time materialization rewrites as expected when the rewritten target exists and is non-empty. For example, `.mcp.json` may use plugin-local or absolute installed paths for CodeGraph/MCP runtimes; that is PASS/WARN context, not payload drift. Missing or zero-byte rewritten targets are FAIL.
+   - Treat install-time materialization rewrites as expected when the rewritten target exists and is non-empty. For example, `.mcp.json` may use plugin-local or absolute installed paths for MCP runtimes; that is PASS/WARN context, not payload drift. Missing or zero-byte rewritten targets are FAIL.
    - Stale project-local leftovers the installer now removes (e.g. `.codex/hooks.json`, `.codex/skills` in the project) are flagged, not deleted.
 4. Probe the real surface. Do not invoke `lazycodex doctor`; this skill is already running inside that doctor workflow, so calling it would recurse. Instead run non-recursive probes directly: `codex --version`, `command -v codex`, the bin-link checks above, config/plugin payload inspections, and a trivial non-interactive Codex invocation that loads the plugin. Use the configured Codex default model for the runtime probe unless the user explicitly passed a model override to the doctor surface; never force a guessed/rejected model such as `gpt-5.5-codex-mini`. Capture stderr verbatim; a clean exit with warnings is WARN, not PASS.
 5. Compare for drift. Where installed manifest-declared bundled files differ from the same files at the installed version, or the latest source removed or renamed something the local config still references, record it with both paths. Do not report expected materialization differences, such as absolute `.mcp.json` runtime paths, as drift when their targets exist and are non-empty.
@@ -96,6 +97,7 @@ sync_latest_source openai/codex "$LAZYCODEX_SOURCE_ROOT/openai-codex-source"
 | Check | Verdict | Evidence |
 | --- | --- | --- |
 | Versions current | PASS/WARN/FAIL | [command output or file:line] |
+| Astra readiness | PASS/WARN | [`codex --version` vs the 0.153.1 floor; `gpt-6-astra` present/absent in `$CODEX_HOME/models_cache.json`; configured root `model`] |
 | config.toml integrity | PASS/WARN/FAIL | [evidence] |
 | Plugin payload wiring | PASS/WARN/FAIL | [evidence] |
 | Bin links / aliases | PASS/WARN/FAIL | [evidence] |
@@ -103,7 +105,7 @@ sync_latest_source openai/codex "$LAZYCODEX_SOURCE_ROOT/openai-codex-source"
 | Drift vs latest source | PASS/WARN/FAIL | [evidence, citing `$LAZYCODEX_SOURCE_ROOT/lazycodex-source` or `$LAZYCODEX_SOURCE_ROOT/openai-codex-source` paths] |
 
 ### Remediations
-1. [Most important fix first: exact command or config edit, and what it resolves.]
+1. [Most important fix first: exact command or config edit, and what it resolves. When Astra readiness is WARN: upgrade Codex to 0.153.1 or newer and relaunch it once so `models_cache.json` picks up `gpt-6-astra`.]
 
 ### Known Issues Matched
 - [issue URL — or "none found"]

package/packages/omo-codex/plugin/skills/lsp-setup/SKILL.md

@@ -1,6 +1,6 @@
 ---
 name: lsp-setup
-description: "Configure a Language Server (LSP) for a specific language so editor/agent tooling — diagnostics, go-to-definition, find-references, rename — works. Use when you need to: configure LSP, lsp setup, set up or install a language server, fix 'no LSP server configured' / 'server not installed', choose between servers (basedpyright vs pyright vs ty vs ruff), or wire .codex/lsp-client.json / .opencode/lsp.json. 언어서버 설정. Routes by file extension to references/<language>/README.md for the exact builtin server, per-OS install commands (macOS/Linux/Windows), config snippets for both config files, initialization options, alternatives, and troubleshooting. Ships scripts: detect-lsp.ts (scan a project for languages + each server's install/config status) and verify-lsp.ts (run a real diagnostics roundtrip). Covers typescript, python, go, rust, c/c++, java, kotlin, c#/razor, swift, ruby, php, dart, elixir, zig, lua, bash, yaml, terraform, haskell, julia."
+description: "Configures a language server so editor/agent tooling (diagnostics, go-to-definition, references, rename) works. Use when a project needs an LSP installed or wired, or a 'no LSP server configured' error appears."
 ---
 
 # LSP Setup

package/packages/omo-codex/plugin/skills/programming/SKILL.md

@@ -1,6 +1,6 @@
 ---
 name: programming
-description: "MUST USE for ANY work on .py .pyi .rs .ts .tsx .mts .cts .go files. One philosophy: strict types, modern stacks (Pydantic v2 / serde+thiserror / Zod / gin+sqlc+pgx+slog), modern toolchains (uv+basedpyright+ruff / cargo+clippy+miri / Bun+Biome+tsc / gofumpt+golangci-lint v2+nilaway+go-race), parse-don't-validate, exhaustive match, typed errors, no any/unwrap/panic, 250 LOC ceiling, TDD, consumer-routed logging. Routes to references/{python,rust,typescript,rust-ub,go}/ + references/logging.md. Triggers: write/edit Python/Rust/TypeScript/Go code, new project, gin server, bubbletea TUI, CJK IME, connect-go RPC, sqlc pgx, branded ids, exhaustive match, unsafe Rust, miri, oversized file, refactor, TDD, e2e test, logging, log levels, structured logging, observability, arena, allocator, bumpalo, const fn, const generics, comptime, zero-alloc, bitfield, repr, scopeguard, errdefer, Zig-like, zerocopy, packed struct."
+description: "Applies strict, modern language practice (typed errors, exhaustive match, TDD) for Python, Rust, TypeScript, and Go. Use for work on .py, .rs, .ts, or .go files."
 ---
 
 # Programming

package/packages/omo-codex/plugin/skills/refactor/SKILL.md

@@ -1,6 +1,6 @@
 ---
 name: refactor
-description: "Intelligent refactor command. Triggers: refactor, refactoring, cleanup, restructure, extract, simplify, modernize."
+description: "Guides a refactor, cleanup, or restructure with the right decomposition. Use when the user asks to refactor, simplify, extract, or modernize code."
 ---
 
 ## Codex Harness Tool Compatibility
@@ -12,17 +12,19 @@ This skill may include examples copied from the OpenCode harness. In Codex, do n
 | `call_omo_agent(subagent_type="explore", ...)` | `multi_agent_v1.spawn_agent({"message":"TASK: act as an explorer. ...","agent_type":"explorer","fork_context":false})` |
 | `call_omo_agent(subagent_type="librarian", ...)` | `multi_agent_v1.spawn_agent({"message":"TASK: act as a librarian. ...","agent_type":"librarian","fork_context":false})` |
 | `task(subagent_type="plan", ...)` | `multi_agent_v1.spawn_agent({"message":"TASK: act as a planning agent. ...","agent_type":"plan","fork_context":false})` |
-| `task(subagent_type="oracle", ...)` for final verification | `multi_agent_v1.spawn_agent({"message":"TASK: act as a rigorous reviewer. ...","agent_type":"lazycodex-gate-reviewer","fork_context":false})` |
-| `task(category="...", ...)` for implementation or QA | `multi_agent_v1.spawn_agent({"message":"TASK: act as an implementation or QA worker. ...","fork_context":false})` |
+| `task(subagent_type="oracle", ...)` for final verification | By default, record a self-review in the notepad: re-read the diff, run diagnostics, and capture evidence for every acceptance criterion. Only when the user demanded strict, rigorous, or high-accuracy review, use `multi_agent_v1.spawn_agent({"message":"TASK: act as a rigorous reviewer. ...","agent_type":"lazycodex-gate-reviewer","fork_context":false})`; include `fork_context: false`. |
+| `task(category="...", ...)` for implementation or QA | `multi_agent_v1.spawn_agent({"message":"TASK: act as an implementation or QA worker. ...","agent_type":"lazycodex-worker-medium","fork_context":false})` |
 | `background_output(task_id="...")` | `multi_agent_v1.wait_agent(...)` for mailbox signals |
 | `team_*(...)` | Use Codex native subagents via `multi_agent_v1.spawn_agent` and `multi_agent_v1.wait_agent`; use `multi_agent_v1.send_input` and `multi_agent_v1.close_agent` only when exposed in the active tools list |
 
-Role-specific behavior must be described in a self-contained `message`. Use `fork_context: false` to start the child with only the initial prompt (no parent history); use `fork_context: true` only when full parent history is truly required. Include any required conversation context, files, diffs, constraints, and requested skill names directly in the spawned agent's `message`. OMO installs these selectable agent roles into `~/.codex/agents/`: `explorer`, `librarian`, `plan`, `momus`, `metis`, `lazycodex-code-reviewer`, `lazycodex-qa-executor`, and `lazycodex-gate-reviewer` - pass the matching name as `agent_type` so the child gets that role's model and instructions. If the spawn tool exposes no `agent_type` parameter, omit it and describe the role inside `message`. If a code block below conflicts with this section, this section wins.
+Role-specific behavior must be described in a self-contained `message`. Use `fork_context: false` to start the child with only the initial prompt (no parent history); use `fork_context: true` only when full parent history is truly required. Include any required conversation context, files, diffs, constraints, and requested skill names directly in the spawned agent's `message`. OMO installs these selectable agent roles into `~/.codex/agents/`: `explorer`, `librarian`, `plan`, `momus`, `metis`, `lazycodex-code-reviewer`, `lazycodex-qa-executor`, and `lazycodex-gate-reviewer` - pass the matching name as `agent_type` so the child gets that role's model and instructions. Inspect the actual spawn schema: whenever `agent_type` is exposed, EVERY spawn MUST select an exact LazyCodex role, on V1 or V2. Implementation difficulty selects `lazycodex-worker-low`, `lazycodex-worker-medium`, or `lazycodex-worker-high`; clone QA can select `lazycodex-clone-fidelity-reviewer`. Never select generic `worker` or `default`. If a code block below conflicts with this section, this section wins.
 
-Codex exposes ONE of two subagent tool surfaces per session; check your own tool list and route accordingly. If `multi_agent_v1.*` tools exist, use the table above as written. If instead a flat `spawn_agent` with a required `task_name` exists (`multi_agent_v2`), rewrite every `multi_agent_v1.*` example: `multi_agent_v1.spawn_agent({...,"fork_context":false})` becomes `spawn_agent({"task_name":"<lowercase_digits_underscores>","message":...,"agent_type":...,"fork_turns":"none"})` (`"all"` only when full parent history is truly required); `send_input` becomes `send_message`; do not call `close_agent`/`resume_agent` (finished agents end on their own; `followup_task` re-tasks one, `interrupt_agent` stops one); `wait_agent` takes only `timeout_ms` and returns on any child mailbox activity. On the v2 surface `agent_type` may be ABSENT from the spawn schema (verified 2026-07-11: only `fork_turns`/`message`/`task_name`) — when absent, omit it and describe the role inside `message`; installed role TOMLs cannot be selected on that surface. If a code block below conflicts with this section, this section wins.
+Codex exposes ONE of two subagent tool surfaces per session; check your own tool list and route accordingly. If `multi_agent_v1.*` tools exist, use the table above as written. If instead a flat `spawn_agent` with a required `task_name` exists (`multi_agent_v2`), rewrite every `multi_agent_v1.*` example: `multi_agent_v1.spawn_agent({...,"fork_context":false})` becomes `spawn_agent({"task_name":"<lowercase_digits_underscores>","message":...,"agent_type":...,"fork_turns":"none"})` (`"all"` only when full parent history is truly required); `send_input` becomes `send_message`; do not call `close_agent`/`resume_agent` (finished agents end on their own; `followup_task` re-tasks one, `interrupt_agent` stops one); `wait_agent` takes only `timeout_ms` and returns on any child mailbox activity. Do not infer role support from V1/V2 or a model version. Legacy-schema exception: only if the actual schema lacks `agent_type`, omit that unsupported field and carry the complete role instructions in `message`, with history explicitly disabled. This cannot select a specialized TOML; an installed managed default supplies the medium worker for unnamed non-forks. The guard has no schema metadata and rejects unnamed calls, so report incompatible routing instead of retrying generically. Every deliberate full-history fork must still name its role: Codex skips role application on unnamed full-history forks, an upstream gap no LazyCodex default can repair. If a code block below conflicts with this section, this section wins. `fork_context` is rejected on `multi_agent_v2` (`fork_context is not supported in MultiAgentV2; use fork_turns instead`).
 
 When translating `load_skills=[...]`, include the requested skill names in the spawned agent's `message`. If a code block below conflicts with this section, this section wins.
 
+Omit optional keys you do not set. Never send `items: []`, `message: ""`, `model: ""`, `reasoning_effort: ""`, or `service_tier: ""` — Codex rejects them (`Items can't be empty`, `reasoning_effort must not be empty`).
+
 For work likely to exceed one wait cycle, require the child to send `WORKING: <task> - <current phase>` before long passes and `BLOCKED: <reason>` only when progress stops. A `multi_agent_v1.wait_agent` timeout only means no new mailbox update arrived; back off between waits (double the timeout up to ~5 minutes) instead of spinning short cycles. Treat a running child as alive. Fallback only when the child is completed without the deliverable, ack-only after followup, explicitly `BLOCKED:`, or no longer running.
 
 export const REFACTOR_TEMPLATE = `# Intelligent Refactor Command
@@ -706,7 +708,7 @@ Record the chosen path in the TodoWrite list.
 
 Rationale for this composition:
 - **4 workers = team mode's parallel cap.** 5+ just queues.
-- **No verifier team member.** Verification needs \`deep\` reasoning (or \`unspecified-high\` fallback). In-team category routing downcasts to sisyphus-junior, which is weaker than required — the verifier runs OUTSIDE the team as a \`task(category="deep")\`.
+- **No verifier team member.** Verification needs \`deep-high\` reasoning (or \`unspecified-high\` fallback). In-team category routing downcasts to the category worker, which is weaker than required — the verifier runs OUTSIDE the team as a \`task(category="deep-high")\`.
 - **quick × 2** for mechanical edits, **unspecified-low × 2** for reasoning edits — mirrors the plan's split.
 
 **Team lifecycle** (one team, reused until Phase 6 cleanup):
@@ -735,10 +737,10 @@ Rationale for this composition:
 While any team task is \`pending | claimed | in_progress\`:
 
 - Wait for \`<system-reminder>\` or member messages. Avoid tight polling; a single \`team_status\` check is acceptable if no notification arrives within roughly 10 seconds of expected completion.
-- On a worker completion report, immediately dispatch an **external verifier** — verification runs OUTSIDE the team because team-member category routing downcasts to sisyphus-junior:
+- On a worker completion report, immediately dispatch an **external verifier** — verification runs OUTSIDE the team because team-member category routing downcasts to the category worker:
   \`\`\`
   task(
-    category="deep",
+    category="deep-high",
     load_skills=[],
     run_in_background=true,
     description="verify step <N>",
@@ -772,5 +774,5 @@ Append to the 6.6 summary a "Dispatch path" line and, when team path was used, t
 - Do not inline the Intent Card or verify-spec into task descriptions — rely on the broadcasts.
 - Do not recreate the team mid-session.
 - Do not run tests from Lead — the external verifier owns that lane.
-- Do not put \`oracle\` / \`librarian\` / \`deep\` into the team spec — oracle/librarian are team-ineligible, and \`deep\` under category routing downcasts to sisyphus-junior. Use them via \`task()\` outside the team when needed.
+- Do not put \`oracle\` / \`librarian\` / \`deep\` into the team spec — curated read-only agents are team-ineligible, and \`deep\` under category routing downcasts to the category worker. Use them via \`task()\` outside the team when needed.
 `