Define an implementation contract for Codex
Turn a feature request into scope, invariants, acceptance criteria, and concrete completion evidence.
354,617 AI resources indexedWorking toward the world’s largest AI resource index.
Feature implementation, debugging, testing, and code review for Codex.
Counts cover every published resource in the catalog. A resource can belong to several fields and platforms.
Turn a feature request into scope, invariants, acceptance criteria, and concrete completion evidence.
Translate an issue report into a reproducible engineering task without inventing missing evidence.
Give Codex a bounded test failure with a behavior contract and limits on implementation changes.
Define a measurable performance question with workload, baseline, and comparison rules.
Draft migration requirements covering mixed versions, existing data, rollback, and validation.
Describe CLI behavior across arguments, streams, exit codes, and machine-readable output.
Frame a review around concrete failure modes, invariants, and affected behavior.
Create a concise repository guide using supplied structure and verified development commands.
Create fixture requirements that isolate behavior and include meaningful boundary cases.
Ask Codex to trace error propagation, cleanup, and caller-visible behavior.
Build one user-visible feature slice while preserving repository conventions and existing work.
Restructure a bounded area while proving important observable behavior stays the same.
Try a broader search or clear some filters.
Platform labels describe intended use. Read each resource's author credits, requirements and available access methods.