Borrowing it
Nothing to install: this file belongs to Coefficient-work/mcp-doctor. 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/Coefficient-work/mcp-doctor/main/.cursor/skills/codex-vs-cursor/SKILL.mdgit clone --depth 1 https://github.com/Coefficient-work/mcp-doctorWrote 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/coefficient-work/mcp-doctor/codex-vs-cursor)<a href="https://agentmods.dev/skills/coefficient-work/mcp-doctor/codex-vs-cursor"><img src="https://agentmods.dev/badge/skills/coefficient-work/mcp-doctor/codex-vs-cursor/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/coefficient-work/mcp-doctor/codex-vs-cursor"><img src="https://agentmods.dev/badge/skills/coefficient-work/mcp-doctor/codex-vs-cursor.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 | $0.00074 | $0.00763 |
| Opus 5.5 | $0.00030 | $0.00305 |
| Sonnet 5.5 | $0.00015 | $0.00153 |
| Haiku 4.5 | $0.00007 | $0.00076 |
Grade A, and why
codex-vs-cursor 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 — 73 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Codex vs Cursor — agent routing
Route the coding agent, not the framework/stack. Do not recommend Claude Code as default. Do not push Hermes, Buzz, Obsidian, or Cloudflare Workers for this decision.
Defaults
| Tool | Role |
|---|---|
| Cursor | Daily IDE — UI work, Flutter, Payload admin, TARS ops, fuzzy explore/debug, tight interactive loops, reviewing diffs live |
| Codex | Walk-away / async agent — scoped migrations, schema PRs, batch fixes, FlutterFlow/Firebase cleanup, large refactors where Louis can leave and come back to a PR |
Decision checklist
Prefer Cursor when any of these dominate:
- Visual / UI iteration (layout, Flutter widgets, Payload admin screens)
- Unclear scope — need explore → poke → adjust in one session
- TARS hub ops, finance scripts, Gmail/remote bridge, Mac mini control
- Live diff review, pair-style debugging, “stay with me while we fix this”
- Small edits that benefit from IDE context (open files, canvas, terminal)
Prefer Codex when most of these are true:
- Scope is writable as a bounded PR (files/areas named up front)
- Work is batchy or mechanical (migrate X→Y, fix N call sites, schema + follow-through)
- Louis wants to walk away and return to a PR / branch
- FlutterFlow or Firebase cleanup / large generated-code sweeps
- Multi-hour refactor where interactive back-and-forth adds little
Split the job when both apply: Cursor for discovery and acceptance criteria → Codex for the bulk PR → Cursor for review polish.
Output format (when advising)
Keep it short:
- Recommendation: Cursor | Codex | Split
- Why: 1–2 bullets from the checklist above
- If Codex: one-line PR/scope brief Louis can paste (goal, in-scope paths, out-of-scope, done when)
- If Cursor: stay in-session; no walk-away framing needed
Repo notes (optional)
| Repo / area | Lean |
|---|---|
jobs-foundation-website |
Codex especially useful (WP→Payload/media migrations, batch content) |
yjordan |
Codex strong for scoped batch/migration PRs |
| Flutter / Firebase batch work | Codex for cleanup sweeps; Cursor for UI polish |
tars, paedia, slj |
Cursor primary (UI/ops, interactive loops) |
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 · 73 lines · 74 tokens per session scan A 7b0df5f74b7c
codex-vs-cursor is a skill published in the GitHub repository Coefficient-work/mcp-doctor (0 stars, last pushed 2mo ago), licensed MIT. It adds 74 tokens to every session and 763 once invoked, about $0.0003 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.
codex-vs-cursor compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| codex-vs-cursorCoefficient-work/mcp-doctor | Skill | 0 | 2mo ago | MIT | A | 74 | Cursor |
| imagegenopenai/codex | Skill | 133k | yesterday | Apache-2.0 | A | 113 | Codex |
| playwright-climicrosoft/playwright | Skill | 99k | yesterday | Apache-2.0 | A | 19 | Claude Code |
| local-ai-agentsmicrosoft/ai-agents-for-beginners | Skill | 79k | 4d ago | MIT | A | 200 | Claude Code, Codex |
| systematic-debuggingobra/superpowers | Skill | 313k | 3d ago | MIT | A | 21 | Claude Code |
| next-cache-components-adoptionvercel/next.js | Skill | 143k | today | MIT | A | 95 | Claude Code |
| next-partial-prefetching-adoptionvercel/next.js | Skill | 143k | today | MIT | A | 103 | Claude Code |