pstack: honor inherit-parent / auto (omit Task model)

[poteto]

Summary

• make inherit-parent and auto valid /setup-pstack values that omit the Task model field
• resolve configured role values across pstack, including Arena's cross-judge pool
• keep Interrogate's closest-slug fallback limited to resolved real slugs
• document Auto-plan behavior in the setup skill and bump pstack to 0.11.5

Testing

• node scripts/validate-plugins.mjs
• git diff HEAD^ --check

  

[cursor]

Cursor Bugbot has reviewed your changes using high effort and found 2 potential issues.

Bugbot Autofix prepared fixes for both issues found in the latest run.

• ✅ Fixed: Cross-judge omits pool config
  • Phase C now selects from the configured cross-judge pool, resolves aliases by omitting model, and permits inherited judges when no different-family entry exists.
• ✅ Fixed: Interrogate fallback breaks inherit
  • The fallback now retries rejected inherit-parent and auto aliases without model instead of substituting an explicit slug.

Or push these changes by commenting:

@cursor push de47c8f430

Preview (de47c8f430)

diff --git a/pstack/skills/arena/SKILL.md b/pstack/skills/arena/SKILL.md
--- a/pstack/skills/arena/SKILL.md
+++ b/pstack/skills/arena/SKILL.md
@@ -38,8 +38,10 @@
 
 ## Phase C: Cross-judge
 
-After all Phase B candidates complete, spawn one readonly judge subagent on a different model family from the parent's. It sees the rubric and the candidates by path label, scores each criterion, and recommends a base with rationale. It runs in parallel with the parent's reading in Phase D, not with the candidates themselves. Spawning while candidates are still writing means the judge sees partial or empty outputs and reports them as dropouts.
+After all Phase B candidates complete, select one judge from your configured `arena cross-judge pool`, preferring an entry from a different model family than the parent's when possible. Resolve the selected entry per `~/.cursor/rules/pstack-models.mdc`. Pass a real slug as `model`; omit `model` for `inherit-parent`/`auto`. If the role line is absent, select from defaults `claude-opus-4-8-thinking-xhigh`, `gpt-5.5-high-fast`, `grok-4.5-fast-xhigh`. Do not require a different family when the pool cannot provide one; an all-alias pool still yields one inherited judge with `model` omitted.
 
+Spawn the judge as a readonly subagent. It sees the rubric and the candidates by path label, scores each criterion, and recommends a base with rationale. It runs in parallel with the parent's reading in Phase D, not with the candidates themselves. Spawning while candidates are still writing means the judge sees partial or empty outputs and reports them as dropouts.
+
 ## Phase D: Pick a base
 
 Read every candidate end to end before picking. Skimming N candidates surfaces only the candidate whose surface looks most familiar.

diff --git a/pstack/skills/interrogate/SKILL.md b/pstack/skills/interrogate/SKILL.md
--- a/pstack/skills/interrogate/SKILL.md
+++ b/pstack/skills/interrogate/SKILL.md
@@ -40,7 +40,7 @@
 - `model`: resolve that `interrogate reviewers` entry per `~/.cursor/rules/pstack-models.mdc`. Pass a real slug as `model`. Omit `model` for `inherit-parent`/`auto`.
 - `readonly`: `true`
 
-If a configured model slug is rejected as unresolvable when you try to spawn the subagent, check the valid slugs in the Task tool's error message, pick the closest equivalent (prefer the highest-reasoning tier of the same family), spawn with the valid slug, and open a separate PR to update the configured defaults. Do not block the review on the slug issue.
+If `inherit-parent` or `auto` is rejected because it was passed as `model`, retry with `model` omitted; never substitute a slug for either alias. If any other configured model slug is rejected as unresolvable when you try to spawn the subagent, check the valid slugs in the Task tool's error message, pick the closest equivalent (prefer the highest-reasoning tier of the same family), spawn with the valid slug, and open a separate PR to update the configured defaults. Do not block the review on the slug issue.
 
 Read `references/reviewer-prompt.md` and fill in the template with:
 1. The stated intent

_{You can send follow-ups to the cloud agent here.}

^{Reviewed by Cursor Bugbot for commit fdf8a27. Configure here.}