mcpField
C
code-review-context

code-review-context

github.com/openaiTesting & quality131.4k450 installsUpdated today
Repository

About code-review-context

A Codex code-review skill for checking changes that affect the model’s conversation context. It focuses on incremental construction, bounded content, cache stability, and required fragment definitions.

Key features

  1. 01Check incremental context construction
  2. 02Avoid cache misses from frequent changes
  3. 03Enforce hard size limits on injected items
  4. 04Flag new items over 1K tokens for review
  5. 05Require ContextualUserFragment structs

Use cases

  1. 01Review changes that add context items
  2. 02Check context fragments for size and structure

Install

Copy the command and run it in a terminal.

$npx skills add https://github.com/openai/codex --skill code-review-context

After installing, type/code-review-contextin Claude Code to confirm it loaded.

Details

From the project README

Codex maintains a context (history of messages) that is sent to the model in inference requests.

  1. No history rewrite - the context must be built up incrementally.
  2. Avoid frequent changes to context that cause cache misses.
  3. No unbounded items - everything injected in the model context must have a bounded size and a hard cap.
  4. No items larger than 10K tokens.
  5. Highlight new individual items that can cross >1k tokens as P0. These need an additional manual review.
  6. All injected fragments must be defined as structs in core/context and implement ContextualUserFragment trait

Comments

Markdown supported