insane-review
July 9, 2026 · View on GitHub
English | 한국어 | 中文 | 日本語 | Español
insane-review
GPT Pro (the Pro reasoning tier of the current flagship — currently GPT-5.6 Sol) has no API. This plugin uses it from inside Claude Code anyway.
GPT Pro lives only in the ChatGPT web app (subscription) — there is no official API, so you can't reach it from Codex CLI, omc ask, or an agent-council API provider. insane-review drives your logged-in ChatGPT web session over CDP: it packs the relevant code with repomix, sends it to Pro, and harvests the review. Zero API cost — it runs on your existing ChatGPT plan.
Quick Start • Why insane-review? • How it works • Features • Tuning & timeouts • Requirements
Quick Start
1. Add the marketplace (once)
/plugin marketplace add https://github.com/fivetaku/gptaku_plugins.git
2. Install
/plugin install insane-review
3. Restart Claude Code
Required for the plugin to load.
4. Prepare the browser bridge (once per machine)
Pro is web-only, so insane-review needs a real, logged-in browser on a debug port:
# launch Comet (or Chrome) with the CDP port, then log into chatgpt.com and pick the Pro reasoning tier
open -a Comet --args --remote-debugging-port=9222
# verify everything is wired up (node/repomix, playwright, pyperclip, CDP browser)
python3 bin/pack_and_ask.py --check-env
5. Run it
/insane-review review the auth flow in src/auth
Or just say "have Pro review this" / "ask GPT Pro about this design" — Claude figures out the target and packs it.
Why insane-review?
- The only way to reach Pro programmatically — no API exists. A logged-in web session driven over CDP is the bridge, and it costs nothing beyond your subscription.
- Claude picks the complete relevant set — you don't have to hand-list files. For reviews it sends full code (no
--compress, which strips function bodies and produces false "looks fine" verdicts) and audits the packed file list so nothing is silently missing. - fail-closed by design — wrong model, unverified login, truncated prompt, an empty pack, or a previous turn's answer are all refused rather than silently sent or saved. Hardened across four rounds of Pro's own self-review (6 → 0 P0s).
- Two roles, one engine — a standalone reviewer when you ask for a fix/review, or a web-only member of agent-council so Pro can debate alongside Codex/Gemini/others.
- Citations you can follow — line numbers are packed in, so Pro's findings come back as
file:lineyou can jump to.
How it works
"have Pro review this" / council member call
↓
Claude selects the COMPLETE relevant file set (full code — no --compress for reviews)
↓
repomix pack (line numbers · secretlint · packed-file-list audit · token count)
↓
CDP-attach the logged-in ChatGPT session
Select Pro effort (flagship auto-follows; currently GPT-5.6 Sol) → re-open menu and VERIFY (mismatch = abort, fail-closed)
↓
Attach pack + prompt → confirm the prompt actually landed in the composer → send
↓
Wait for THIS turn to complete (turn-scoped: new assistant node + new copy button)
Optionally cut long reasoning early with --force-answer-after
↓
Harvest the answer → save to ./.insane-review/response_*.md (atomic write)
Output lands in the current project's .insane-review/ folder (like kkirikkiri's .kkirikkiri/), never inside the plugin:
.insane-review/
├── pack_<target>_<ts>.md # what was sent (chmod 600)
└── response_<target>_<ts>.md # Pro's answer + verified model header
Features
Commands
| Command | What it does |
|---|---|
/insane-review [target/question] | Pack the relevant code and send it to GPT Pro for review |
| natural language | "have Pro review this", "ask GPT Pro about X" — same flow |
Two modes
- Standalone reviewer — you ask for a fix/review → Claude scopes the target → repomix pack → Pro analysis → applied back.
- agent-council member — register Pro as a web-only council member so it debates with other models. See
references/council-setup.md.
Key flags
| Flag | Purpose |
|---|---|
--target <dir> | Folder to pack (omit for a prompt-only opinion) |
--include <glob> / --ignore <glob> | Narrow the packed set |
--model pro | Select the reasoning effort (e.g. Pro) |
--require-model "GPT-5.6" | Verify the active model name — abort send on mismatch (fail-closed) |
--prompt "..." / --prompt-file | The question |
--pack-only | Just pack (inspect token count), don't send |
--council | Council mode — response on stdout, logs on stderr |
--compress | tree-sitter skeleton only — don't use for reviews (drops function bodies) |
--check-env / --install | Diagnose / install the local toolchain |
Tuning & timeouts
Response waiting and pack timeouts are adjustable from both the CLI and the environment — useful because Pro's full reasoning can take 10–15 minutes.
| Control | Default | What it does |
|---|---|---|
--max-wait <sec> | 1200 (20 min) | Max time to wait for Pro's response before giving up (fail-closed, no partial save) |
INSANE_REVIEW_MAX_WAIT | 1200 | Same as --max-wait, via environment |
--force-answer-after <sec> | off | Soft cut: if Pro is still reasoning after N sec, click "Get answer now" so it answers from the reasoning it has done so far — a complete, saved answer (see below) |
INSANE_REVIEW_REPOMIX_TIMEOUT | 300 | Max seconds for the repomix packing step |
--retries <n> | 1 | Re-attempts if a send/harvest fails |
Two different "timeouts" — don't confuse them:
--force-answer-after N(soft cut, recommended for bounding cost). Pro reasons for a long time; this clicks ChatGPT's "Get answer now" at N seconds, so Pro stops reasoning and replies based on what it has reasoned up to that point. That reply is a normal, complete turn — it's harvested and saved like any other. Use it to cap a council member at, say, 120s instead of waiting 10+ minutes.--max-wait N(hard ceiling, fail-closed). If the turn never completes within N seconds and no answer was forced, insane-review gives up without saving the half-streamed text — an incomplete answer is treated as a failure, not a result. This is intentional: it never hands you a truncated review pretending to be done.
Other environment overrides:
| Variable | Default | What it does |
|---|---|---|
INSANE_REVIEW_CDP_PORT | 9222 | Browser remote-debugging port |
INSANE_REVIEW_COMET / INSANE_REVIEW_CHROME | app default path | Browser executable path |
INSANE_REVIEW_REPOMIX_VERSION | 1.15.0 | Pinned repomix version (reproducibility) |
INSANE_REVIEW_OUT | ./.insane-review | Output directory (also --out-dir) |
# example: give Pro up to 25 minutes, but cut reasoning at 5 minutes if it's still thinking
INSANE_REVIEW_MAX_WAIT=1500 python3 bin/pack_and_ask.py \
--target . --include "src/**" --model pro --require-model "GPT-5.6" \
--force-answer-after 300 --prompt "Where are the concurrency bugs?"
Requirements
Required
- Claude Code
- Python 3.11+ with
playwrightandpyperclip - Node.js /
npx - A subscription ChatGPT account with GPT Pro, logged in inside Comet/Chrome launched on the debug port (
--remote-debugging-port=9222)
What's auto-handled vs. what you do
| Dependency | First-run behavior |
|---|---|
| repomix | Fully automatic — pulled on demand via npx -y repomix@<pinned>, never needs manual install |
| playwright / pyperclip | Checked on first use by --check-env; install them with --install (runs pip install). A normal run without them stops with a clear instruction (fail-closed) rather than failing mid-way |
| Browser login + GPT Pro | Manual — can't be automated; you log into chatgpt.com and select Pro once |
# one shot: checks node/repomix, playwright, pyperclip, CDP browser — and installs the pip deps if missing
python3 bin/pack_and_ask.py --check-env --install
Note
Web-UI automation is not endorsed by OpenAI's ToS and selectors may need maintenance when the ChatGPT DOM changes. Intended for personal subscription use only.
License
MIT
No API. Still Pro.