Borrowing it
Nothing to install: this file belongs to ifokeev/airbrowser-mcp. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/ifokeev/airbrowser-mcp/master/.claude/skills/codex-review/SKILL.mdgit clone --depth 1 https://github.com/ifokeev/airbrowser-mcpWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/ifokeev/airbrowser-mcp/codex-review)<a href="https://agentmods.dev/skills/ifokeev/airbrowser-mcp/codex-review"><img src="https://agentmods.dev/badge/skills/ifokeev/airbrowser-mcp/codex-review/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/ifokeev/airbrowser-mcp/codex-review"><img src="https://agentmods.dev/badge/skills/ifokeev/airbrowser-mcp/codex-review.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00037 | $0.00919 |
| Opus 5.5 | $0.00015 | $0.00368 |
| Sonnet 5.5 | $0.00007 | $0.00184 |
| Haiku 4.5 | $0.00004 | $0.00092 |
Grade A, and why
codex-review scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2mo ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 118 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Codex Code Review
Run OpenAI Codex CLI to review code changes or implementation plans.
Mode Detection
Determine which mode to use based on what the user wants reviewed:
Mode A — Code Diff Review (codex review)
Use when the user wants to review actual code changes (committed or uncommitted).
Indicators: "review my changes", "review this PR", "review the diff", or no specific context (default).
Mode B — Plan / Design Review (codex exec)
Use when the user wants to review a plan, design, or suggested implementation that exists in conversation context — NOT as code diffs.
Indicators: "review the suggested implementation", "review this plan", "review the design", "review what was proposed", or when there are no relevant code changes on the branch.
Compact Prompt
Always pass a compact-output instruction via heredoc stdin to reduce context window usage. codex review does NOT accept a prompt as a positional argument alongside flags, but it does read from stdin via heredoc.
COMPACT_PROMPT="Be concise. For each finding output ONLY: severity (critical/warning/info), file:line, and a one-line description. Group by severity. No code snippets unless critical. No preamble or summary paragraph."
Mode A — Code Diff Review
-
Run
git statusto check for uncommitted changes -
Choose the appropriate review target:
Uncommitted changes (staged, unstaged, or untracked):
codex review --uncommitted <<'EOF' $COMPACT_PROMPT EOFSpecific commit SHA:
codex review --commit <SHA> <<'EOF' $COMPACT_PROMPT EOFBase branch comparison:
codex review --base <BRANCH> <<'EOF' $COMPACT_PROMPT EOFNo uncommitted changes — review latest commit:
codex review --commit HEAD <<'EOF' $COMPACT_PROMPT EOF -
Present the output and offer to help fix issues
Mode B — Plan / Design Review
Use codex exec in read-only sandbox mode. Pipe the plan text via stdin.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 2mo ago First seen · 118 lines · 37 tokens per session scan A babf36c1dadb
codex-review is a skill published in the GitHub repository ifokeev/airbrowser-mcp (10 stars, last pushed 4mo ago), licensed MIT. It adds 37 tokens to every session and 919 once invoked, about $0.0001 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
composing-evidence
Use when a live page, or a screenshot attached to a ticket, needs to become polished evidence for an MR or ticket, such as proving a value changed, showing a fixed bug, or marking what a reviewer should look at, and the result should be a cropped, labelled image rather than a marked-up full screenshot.
ego-lite-simplify
Find and implement evidence-backed simplifications in the ego-lite repository. Use when reviewing the codebase for dead code, duplicated state or APIs, speculative abstractions, unnecessary compatibility layers, hand-written infrastructure, excessive tests or documentation, or when the user asks to reduce code size or…
surf
Drive the user's logged-in Chrome from the shell. Use for browser automation (open, read, click, fill, upload, screenshot), reading a site's feed or API, or asking ChatGPT, Gemini, Perplexity or Grok through the browser.
deep-x-research
Deep, exhaustive research on a topic across X (Twitter) by driving Grok (x.com/i/grok) through surf. Use when the user wants comprehensive X research on a concept, technique, trend, tool, or creator scene; needs categorized findings with every claim traceable to post URLs; or when a single Grok query is not enough.
bladebro
Optional house rules for bladebro. Bladebro itself needs no skill — the package registers its MCP server with pi and the five tools arrive complete. Add your own preferences here if you want the agent to use bladebro in a particular way.
code-quality
Check existing-project implementation consistency, simplify recent code changes without altering behavior, or review changes for bugs and regressions. Choose the relevant section by task; review-only requests must not modify files.