max-metaplanet · Nov 6, 2026 · Nov 6, 2026 · Nov 6, 2026
diff --git a/CHANGES.md b/CHANGES.md
@@ -3,6 +3,10 @@
 Open Pstack lands on green: Mergify merges a ready PR to main once verify and exact-head live-gate pass and no review thread is unresolved, and a workflow resolves Unfret's own threads. A PR that touches `.github/` or the Mergify config also waits for Unfret, and workflow changes require operator queueing.
 This port applies the Cursor → Claude Code substitutions in skill bodies. Earlier drafts left them flagged; this revision resolves them. A later pass added a Codex build that shares the same skills; see [Codex port](#codex-port) below.
 
+## Unreleased
+
+**Grok is a parent harness.** A Grok session resolves routes from the catalog's Grok parent column. A Grok lane whose model and effort already match the parent uses `spawn_subagent`. Every other Grok lane, and every Claude or Codex lane, uses the runner with `--parent grok`. The runner accepts a Grok provider from a Grok parent and sets `--permission-mode auto`. Its help says a Claude or Codex parent still rejects its own provider. Claude-on-Claude and Codex-on-Codex stay on the native route. A headless Grok parent stays in auto mode, so a permission prompt reports the blocked call instead of cancelling the turn.
+
 ## 1.20.0 gives Amp an opted-in writer
 
 Open Pstack 1.20.0 keeps the Cursor pstack 0.15.36 sync. Default roles and panels are unchanged.

diff --git a/README.md b/README.md
@@ -18,7 +18,7 @@ Compared with Lauren's pstack, Open Pstack adds:
 
 - **A model and provider registry.** Models, efforts and providers live in one typed [registry](plugins/pstack/skills/poteto-mode/scripts/runner/registry/). The [model catalog](plugins/pstack/skills/poteto-mode/references/model-catalog.md) and the Claude-native agent files are generated from it, and the skills point at the catalog's role defaults instead of naming models. A new model is one registry entry. A new command-line provider is one adapter module and one line in the provider list. See the [design](docs/plans/2026-10-11-model-provider-registry.md).
 - **One skill tree for Claude Code and Codex.** Both apps read the same skills, and one [reference](plugins/pstack/skills/poteto-mode/references/codex-tools.md) maps how each app starts them. Lauren's pstack does not run in Codex. See [Claude Code and Codex](#claude-code-and-codex) below.
-- **Models from other providers.** Claude Code and Codex hand work to other providers' models through each provider's own command-line tool, started by one [runner](plugins/pstack/skills/poteto-mode/references/provider-dispatch.md). The runner supports nine providers. Claude, Codex and Grok run lanes that read or write files. Devin, DeepSeek Harness, Antigravity, OpenCode, Amp and Cursor CLI run read-only lanes. The model catalog's [routes table](plugins/pstack/skills/poteto-mode/references/model-catalog.md#routes) lists how each one starts. Each run leaves a receipt with the exact model flags and, where the provider reports it, the model that answered. Every provider except Claude and Codex gets only an allowlisted environment. A Claude lane that writes files runs in Claude Code's Bash sandbox, with writes to the shared Git directory denied. A model that cannot run fails visibly; the runner never switches to a weaker model.
+- **Models from other providers.** Claude Code and Codex hand work to other providers' models through each provider's own command-line tool, started by one [runner](plugins/pstack/skills/poteto-mode/references/provider-dispatch.md). The runner supports nine providers. Claude, Codex and Grok run lanes that read or write files. Devin, DeepSeek Harness, Antigravity, OpenCode, Amp and Cursor CLI run read-only lanes. The model catalog's [routes table](plugins/pstack/skills/poteto-mode/references/model-catalog.md#routes) lists how each one starts. A Grok parent runs a Grok lane whose model and effort already match the parent with `spawn_subagent`, and every other lane through the runner with `--parent grok`. Each run leaves a receipt with the exact model flags and, where the provider reports it, the model that answered. Every provider except Claude and Codex gets only an allowlisted environment. A Claude lane that writes files runs in Claude Code's Bash sandbox, with writes to the shared Git directory denied. A model that cannot run fails visibly; the runner never switches to a weaker model.
 - **Runs inside T3 Code.** In [T3 Code](plugins/pstack/skills/poteto-mode/references/provider-dispatch.md#t3-host), pstack keeps each role's model and effort. T3 starts each lane whose provider it offers, and the runner starts the rest.
 - **A checked model sheet.** [`setup-pstack`](plugins/pstack/skills/setup-pstack/SKILL.md) records each role as `provider:model@effort`, asks for one effort per model family, and probes each chosen model before it saves anything. A rerun that changes nothing keeps the sheet without asking again.
 - **Three-model panels.** `arena`, `architect` and `interrogate` default to Opus, Sol and Grok. Lauren's pstack defaults to Opus 5.5 and Grok; Open Pstack keeps three ([UPSTREAM.md](UPSTREAM.md)).

diff --git a/UPSTREAM.md b/UPSTREAM.md
@@ -71,6 +71,8 @@ The table above is the current Cursor sync point. Open Pstack 1.17.0 imported th
 
 - **#253 — T3 Code parent**: Cursor has no T3 Code route. The port adds the `T3 host` section to `provider-dispatch.md`: a lane whose provider T3 offers and has enabled goes through `delegate_task`, and any other lane takes the standalone route of the Claude Code or Codex runtime. `codex-tools.md` points to it, and `setup-pstack/SKILL.md` routes its probes and smoke panel the same way at its existing steps. No other skill body changes. Keep these lines during syncs.
 
+- **#281 — Grok parent**: Cursor has no Grok parent. The port adds `grok` to `PARENTS` with `externalSelf`, so the generated route table has a Grok parent column and the runner accepts `--parent grok`, including a Grok provider. Claude and Codex stay native-only for their own provider. `provider-dispatch.md`, `codex-tools.md`, and `setup-pstack/SKILL.md` name `spawn_subagent` and that runner exception. Keep these lines during syncs.
+
 ## Check for changes
 
 The repository already names Cursor's repository as the `cursor` remote in the maintainer checkout. A fresh clone can add it once:

diff --git a/docs/reference.md b/docs/reference.md
@@ -86,7 +86,7 @@ The Codex build shares one `skills/` tree with the Claude Code build. Nothing is
 - **Tool and built-in mapping.** Claude tool names and built-in skills resolve through [`codex-tools.md`](../plugins/pstack/skills/poteto-mode/references/codex-tools.md). Model execution resolves separately through [`provider-dispatch.md`](../plugins/pstack/skills/poteto-mode/references/provider-dispatch.md), so Codex models stay native while other providers run through the external runner.
 - **Subagents.** The `Agent` tool maps to Codex `spawn_agent` / `wait_agent`, enabled by `multi_agent = true`. Parallel fan-out is multiple `spawn_agent` calls in one turn. If the native Codex lane is unavailable, record that lane as a dropout; the configured external lanes still run, and no provider is silently substituted. There is no `poteto-agent` subagent type on Codex; route ad-hoc subagents by dispatching a `spawn_agent` told to read `poteto-mode` first.
 - **Auto-fire.** The `hooks/` SessionStart injection is Claude Code-only; Codex has no plugin hook runtime. Enter `pstack:poteto-mode` by name, or add a standing instruction to `~/.codex/AGENTS.md` if you want the same always-on routing.
-- **Models.** `/setup-pstack` writes provider-qualified descriptors and asks one requested effort per assigned family (`low`, `medium`, `high`, `xhigh`, `max`, and `ultra` where the model's CLI lists it, today Sol and Astra). The first-run sheet is the model catalog's [role defaults](../plugins/pstack/skills/poteto-mode/references/model-catalog.md#role-defaults), with a panel of Opus, Sol, and Grok; Fable, Sonnet, Luna, Astra, Haiku 5.5 and an exact Opus pin stay selectable. Fable, Opus and Sonnet use Claude's rolling aliases; a registered exact pin keeps its own model. Runtime dispatch normalizes older versioned descriptors in memory, so an installed sheet stops pinning immediately. A setup rerun persists that migration while keeping each role's family and effort. In Codex, Codex families such as Sol use native `spawn_agent`; every other provider uses the deterministic external runner. In Claude Code, Claude families use native agents; every other provider uses the runner. A Codex writer that the runner launches stays in Codex's `workspace-write` sandbox unless the sheet opts in with the line `codex writer sandbox: full-access`, which setup asks about once. Full access lets the writer launch a browser and write dot-folders such as `.agents/`, and removes every sandbox limit. A provider whose writes no sandbox confines, such as OpenCode, Cursor or Muse, writes only when the sheet holds `<provider> writer sandbox: full-access`, which setup asks about when a writer role uses that provider. Children never detect the parent or reroute themselves. The `bug-fix`, `perf-issue`, and `hillclimb` roles stay on Sol instead of upstream's Grok default.
+- **Models.** `/setup-pstack` writes provider-qualified descriptors and asks one requested effort per assigned family (`low`, `medium`, `high`, `xhigh`, `max`, and `ultra` where the model's CLI lists it, today Sol and Astra). The first-run sheet is the model catalog's [role defaults](../plugins/pstack/skills/poteto-mode/references/model-catalog.md#role-defaults), with a panel of Opus, Sol, and Grok; Fable, Sonnet, Luna, Astra, Haiku 5.5 and an exact Opus pin stay selectable. Fable, Opus and Sonnet use Claude's rolling aliases; a registered exact pin keeps its own model. Runtime dispatch normalizes older versioned descriptors in memory, so an installed sheet stops pinning immediately. A setup rerun persists that migration while keeping each role's family and effort. In Codex, Codex families such as Sol use native `spawn_agent`; every other provider uses the deterministic external runner. In Claude Code, Claude families use native agents; every other provider uses the runner. In Grok, a Grok descriptor whose model and effort already match the parent uses `spawn_subagent`. Every other provider, and any other Grok descriptor, uses the runner with `--parent grok`. A Codex writer that the runner launches stays in Codex's `workspace-write` sandbox unless the sheet opts in with the line `codex writer sandbox: full-access`, which setup asks about once. Full access lets the writer launch a browser and write dot-folders such as `.agents/`, and removes every sandbox limit. A provider whose writes no sandbox confines, such as OpenCode, Cursor or Muse, writes only when the sheet holds `<provider> writer sandbox: full-access`, which setup asks about when a writer role uses that provider. Children never detect the parent or reroute themselves. The `bug-fix`, `perf-issue`, and `hillclimb` roles stay on Sol instead of upstream's Grok default.
 
 Verified in fresh installed Claude Code and Codex sessions: the user-facing skills are discovered and namespaced under `pstack`; both parents fan out the configured panel through the documented native/external route table, retain long-running handles without a default timeout, and cross-judge only after every candidate is terminal. The `principle-*` leaves remain available for `poteto-mode` to read by path. Claude honors their `user-invocable: false` metadata; Codex 0.149.0 does not ([#8](https://github.com/ericlitman/open-pstack/issues/8)).
 
@@ -113,7 +113,7 @@ Not declared as deps, but referenced in skill bodies:
 - **Bash on Windows** — the SessionStart hook runs through `hooks/run-hook.cmd`, which looks for Git for Windows' `bash.exe` in its standard system-wide and per-user locations, then for a `bash` on `PATH` (MSYS2, Cygwin). Without one, the hook exits 0 and skips the pstack startup context, as the superpowers wrapper it comes from does. Install [Git for Windows](https://git-scm.com/download/win) to get the context. In Codex, the hook starts `C:\Windows\System32\cmd.exe` by full path, because one command must work in both PowerShell and cmd. A Windows installed outside `C:\Windows` does not run the hook.
 - **macOS, Linux or WSL2 for Claude writer lanes** — Claude Code's sandbox does not run on native Windows, so a Claude writer lane needs one of these hosts.
 - **`node`** — runs `skills/poteto-mode/scripts/check-plan.mjs`. The checker uses only Node built-ins and does not need Bun.
-- **Provider CLIs** — the external runner uses the assigned subscribed CLI directly, one per provider in the [model catalog's route table](../plugins/pstack/skills/poteto-mode/references/model-catalog.md#routes). Install and authenticate only the providers present in your model sheet. Some providers run read-only lanes only. The catalog notes for OpenCode, Amp, Cursor, Devin, agy and dsh say what repository and host configuration their CLIs load into a lane. Same-provider work stays native; the runner refuses it.
+- **Provider CLIs** — the external runner uses the assigned subscribed CLI directly, one per provider in the [model catalog's route table](../plugins/pstack/skills/poteto-mode/references/model-catalog.md#routes). Install and authenticate only the providers present in your model sheet. Some providers run read-only lanes only. The catalog notes for OpenCode, Amp, Cursor, Devin, agy and dsh say what repository and host configuration their CLIs load into a lane. A Claude or Codex provider on its own parent stays native, and the runner refuses that call. A Grok parent may run a Grok provider through the runner.
 - **`jq` and `rg` (ripgrep)** — only for `scripts/worktree-audit.sh` (the Worktree cleanup playbook). Without them the audit still runs but blanks its PR and LAST_CHAT columns, so it warns on stderr rather than returning a table that looks complete.
 
 No third-party plugins. The harsher-critique escape hatch lives in the bundled `thermo-nuclear-code-quality-review` skill (imported from cursor-team-kit), not in an external plugin.
@@ -206,7 +206,7 @@ The port is editorial, not mechanical. Anywhere upstream pstack assumed Cursor-s
 | Cursor cloud agents (`environment: "cloud"`, `cloud_base_branch`) | Local background subagents (`run_in_background: true`), isolated by git worktree |
 | Orchestrate's Project tracker (`notes.md`) and Project store | `<config-home>/orchestrate/<project-slug>/notes.md`, which survives the session restarts a multi-day program expects |
 | Model rule `~/.cursor/rules/pstack-models.mdc` | Override sheet `~/.claude/pstack-models.md`, included from `CLAUDE.md` |
-| Multi-model panels (arena, architect, interrogate) | Provider dispatch runs Open Pstack's default three-model panel of Opus, Sol, and Grok from the model catalog's [role defaults](../plugins/pstack/skills/poteto-mode/references/model-catalog.md#role-defaults). Same-provider lanes stay native; external lanes use the bundled runner. Lauren's pstack now defaults to Opus 5.5 and Grok. |
+| Multi-model panels (arena, architect, interrogate) | Provider dispatch runs Open Pstack's default three-model panel of Opus, Sol, and Grok from the model catalog's [role defaults](../plugins/pstack/skills/poteto-mode/references/model-catalog.md#role-defaults). A Claude or Codex lane on its own parent stays native. A Grok parent may run a Grok lane through the runner. Every other lane uses the bundled runner. Lauren's pstack now defaults to Opus 5.5 and Grok. |
 
 ### Cross-vendor dispatch
 

diff --git a/plugins/pstack/skills/poteto-mode/SKILL.md b/plugins/pstack/skills/poteto-mode/SKILL.md
@@ -9,7 +9,7 @@ Bundled skills live at `../<name>/SKILL.md` from this file, so put this file's a
 
 ## Platform Adaptation
 
-These skills share one tree across Claude Code and Codex. Read [`references/provider-dispatch.md`](references/provider-dispatch.md) whenever a configured role launches. It defines the provider-qualified model descriptors, native/external route table, launcher, isolation, receipts, and dropout policy. Children never choose routes. When a skill names a Claude tool or built-in skill (`run`, `verify`, `plugin-dev:skill-development`), read [`references/codex-tools.md`](references/codex-tools.md) for the Codex equivalent. On Claude Code, read its [Waits on Claude Code](references/codex-tools.md#waits-on-claude-code) section before any wait.
+These skills share one tree across Claude Code, Codex, and Grok. Read [`references/provider-dispatch.md`](references/provider-dispatch.md) whenever a configured role launches. It defines the provider-qualified model descriptors, native/external route table, launcher, isolation, receipts, and dropout policy. Children never choose routes. When a skill names a Claude tool or built-in skill (`run`, `verify`, `plugin-dev:skill-development`), read [`references/codex-tools.md`](references/codex-tools.md) for the Codex equivalent. On Claude Code, read its [Waits on Claude Code](references/codex-tools.md#waits-on-claude-code) section before any wait.
 
 ## Non-negotiables
 
@@ -92,7 +92,7 @@ Read the leaf skill in full for any principle you apply. Each entry names when i
 
 ## Subagents
 
-For `inherit-parent`, `auto`, or an unconfigured native ad-hoc helper, prefer `poteto-agent`. The chat that receives `/poteto-mode` runs the playbook. Never spawn a `poteto-agent` to run the playbook itself. A provider-qualified role instead follows provider dispatch: Claude's shipped frontier agent definitions select the model alias and requested effort, Codex passes both to `spawn_agent`, and external providers run through the deterministic launcher. Routed workflow skills set the task and access mode. Do not override their choices.
+For `inherit-parent`, `auto`, or an unconfigured native ad-hoc helper, prefer `poteto-agent`. The chat that receives `/poteto-mode` runs the playbook. Never spawn a `poteto-agent` to run the playbook itself. A provider-qualified role instead follows provider dispatch: Claude's shipped frontier agent definitions select the model alias and requested effort, Codex passes both to `spawn_agent`, and Grok calls `spawn_subagent` when the descriptor's model and effort are already the parent's. Every other provider, and any other Grok descriptor, runs through the deterministic launcher. Routed workflow skills set the task and access mode. Do not override their choices.
 
 **Defaults for every delegation.** Start independent lanes together. In the main session, run a steer or resume of a running agent in the background too, because a foreground call holds your turn while other reports wait; a Claude Code subagent waits as [Waits on Claude Code](references/codex-tools.md#waits-on-claude-code) says. Use file pointers rather than inlined dumps, preserve only the tools or MCPs the task needs, and assign every writer a worktree or unique output directory. `/setup-pstack` configures the descriptor per role. A role without a sheet line uses its [role default](references/model-catalog.md#role-defaults); the panel roles default to one model from each of three providers. Each code playbook's configured model comes from its line (`feature, refactoring`, `bug-fix`, `perf-issue`, or `hillclimb`), and the hardest changes read `hardest tasks`. Prose and judgment read `judgment and prose`. `inherit-parent` and `auto` use the parent model natively and reduce provider diversity when used in a panel.