OrangeTheWeak
Community@orange-the-weak
Routes Codex project work across GPT-5.6 Sol, Terra, and Luna with bounded concurrency, reasoning-effort control, reviews, tests, and model recommendations.
Agent Skills by OrangeTheWeak
Showing 1 vetted skills indexed across 1 GitHub repositories.
Frequently Asked Questions About OrangeTheWeak
FAQPage SchemaWhich Codex tasks fit this skill?▼
It supports code changes, tests, reviews, routed implementation, model recommendations, usage queries, retuning, and requests to disable, exit, restore, or check status for the current project. It never auto-selects Ultra and never creates a new top-level Codex task.
Who should use orange-the-weak's codex-auto-model-router?▼
Codex users who want route recommendations among GPT-5.6 Sol, Terra, and Luna, plus reasoning-effort guidance for project work. It fits engineers comparing route fit against startup and aggregation overhead before edits, reviews, or test runs.
How does it behave during a Codex session?▼
It recommends and executes an efficient Sol, Terra, or Luna route, prefers bounded direct concurrency, and creates or reuses a route-specific leaf only when benefit clearly exceeds overhead. Strict ledger auditing runs only when explicitly requested.
Are installation steps or dependencies documented?▼
No installation steps, package dependencies, license terms, or external prerequisites are stated in the manifest. The described context is Codex project work, with route selection, reasoning control, status checks, retuning, disable, exit, and restore handling.