ax-check.com
by Gauge
C
CodeRabbit
coderabbit.ai
ax-check.com
Checked 6 October 2026
B
84/100
Agents need your login
CodeRabbit's quickstart and pricing are clear and quick to find.
Docs give a concrete sign-up-and-scan setup with install commands; pricing lists exact per-developer rates ($24/$48/$72, Enterprise custom) right on the page.
2 fixes to start with
Concrete changes, backed by what we found.
Recorded coding sessions
Same task, 3 models, no coderabbit.ai account. Recorded 6 October 2026 on Pi.
3 recorded · 33s
33s · recorded
Session findings
Select a finding to see it in context.
›Thinking…
The user is asking me to build a simple example using CodeRabbit. Let me check the environment — is the CodeRabbit SDK/API available? CodeRabbit is a code review SaaS. To integrate, you'd need an API key and webhooks to GitHub/GitLab, and without credentials you can't operate on a hosted product.
First, let me check whether there are any related tools or credentials in the environment.
ls -la /sandbox/repo && env | grep -iE 'coderabbit|cr_|github|gitlab|api_key' | sed 's/=.*/=<redacted>/' | head -20
total 16
total 16 drwxr-xr-x 3 root root 4096 Sep 18 08:00 . drwxr-xr-x 7 root root 4096 Oct 6 16:36 .. drwxr-xr-x 8 root root 4096 Sep 24 20:05 .git -rw-r--r-- 1 root root 408 Sep 18 08:00 .gitignore PI_GATEWAY_API_KEY=<redacted>
Recorded tool call
{
"type": "tool_call",
"seq": 3,
"ts": "2026-10-06T16:36:55.131Z",
"name": "bash",
"turn": 1,
"input": {
"command": "ls -la /sandbox/repo && env | grep -iE 'coderabbit|cr_|github|gitlab|api_key' | sed 's/=.*/=<redacted>/' | head -20"
},
"run_id": "cmuwwigqd01fk0ivpotoe66fu",
"tool_id": "chatcmpl-tool-be3faf46b7b5850f",
"tool_kind": "shell"
}Recorded tool result
{
"type": "tool_result",
"seq": 5,
"ts": "2026-10-06T16:36:55.131Z",
"turn": 1,
"output": {
"content": [
{
"text": "total 16\ndrwxr-xr-x 3 root root 4096 Sep 18 08:00 .\ndrwxr-xr-x 7 root root 4096 Oct 6 16:36 ..\ndrwxr-xr-x 8 root root 4096 Sep 24 20:05 .git\n-rw-r--r-- 1 root root 408 Sep 18 08:00 .gitignore\nPI_GATEWAY_API_KEY=<redacted>\n",
"type": "text"
}
]
},
"run_id": "cmuwwigqd01fk0ivpotoe66fu",
"tool_id": "chatcmpl-tool-be3faf46b7b5850f",
"is_error": false,
"tool_kind": "shell"
}which coderabbit 2>/dev/null; npm ls -g 2>/dev/null | grep -i coderabbit; pip list 2>/dev/null | grep -i coderabbit
Failed(no output)
(no output) Command exited with code 1
Recorded tool call
{
"type": "tool_call",
"seq": 4,
"ts": "2026-10-06T16:36:55.131Z",
"name": "bash",
"turn": 1,
"input": {
"command": "which coderabbit 2>/dev/null; npm ls -g 2>/dev/null | grep -i coderabbit; pip list 2>/dev/null | grep -i coderabbit"
},
"run_id": "cmuwwigqd01fk0ivpotoe66fu",
"tool_id": "chatcmpl-tool-88bf3740dc2b5b54",
"tool_kind": "shell"
}Recorded tool result
{
"type": "tool_result",
"seq": 6,
"ts": "2026-10-06T16:36:56.938Z",
"turn": 1,
"output": {
"content": [
{
"text": "(no output)\n\nCommand exited with code 1",
"type": "text"
}
],
"details": {}
},
"run_id": "cmuwwigqd01fk0ivpotoe66fu",
"tool_id": "chatcmpl-tool-88bf3740dc2b5b54",
"is_error": true,
"tool_kind": "shell"
}›Thinking…
There are no CodeRabbit credentials in the environment, no SDK/CLI installed, and the repo is essentially empty (just a .gitignore). CodeRabbit is a SaaS for PR reviews that works through a GitHub App / GitLab integration, requires connecting a real repository via the hosted dashboard, and reviews are triggered by PRs. A public REST API with an API key you can just call right away is not provided (there are APIs for some integrations, but they require an account and an installed app). Without credentials/account access, I can't act on the hosted product.
The user's request is: "Tell me how pricing works, and briefly tell me whether this product will be easy for you to manage. Let me know if you're blocked. If this product has no developer workflow you can act on, say so plainly and stop."
I should say this: I'm blocked — no API key/account, and CodeRabbit's workflow is a GitHub-App-based PR review, which requires installing on a real repo and a paid account. I can't build a working example without credentials. I should be upfront and stop. But can I still answer the pricing and manageability questions and possibly provide a config file example (.coderabbit.yaml) that doesn't require a running service? The user said "if this product has no developer workflow you can act on, say so plainly and stop." The core request is "build a simple example" — I can't execute against the hosted product. I should state that plainly, provide pricing info and a manageability assessment, and possibly leave a minimal .coderabbit.yaml as an artifact of an executable workflow (a config file committed to the repo is the actual developer workflow, it becomes active once the app is installed, and no local stack is needed). That seems reasonable and light. But does that qualify as "acting on a developer workflow"? Adding a config file is harmless and useful. Still, I shouldn't over-engineer. I'll add a minimal .coderabbit.yaml and clearly state that without credentials I can't execute anything against the hosted service.
Pricing (as of my knowledge): Free plan for open source and personal use? CodeRabbit pricing: Lite (free), Pro ($12/month/seat for individuals, $24/month/seat Pro for teams?). Actually: Lite — free, Pro for individual developers $12/seat/month, Pro for teams $24/seat/month, Enterprise is custom. Free trial of Pro for 14 days. I should note this may have changed and point them to coderabbit.ai/pricing.
Manageability: Reasonably easy for an agent to manage config (.coderabbit.yaml, path instructions) via git. But triggering and viewing reviews requires access to the GitHub/GitLab PR UI/API — I could interact via the GitHub API with a token, but without a token I'm blocked. I'll state that.
I checked the environment: the repo is essentially empty (just a .gitignore), there's no CodeRabbit CLI/SDK installed, and no CodeRabbit API key or GitHub/GitLab credentials are available.
Plainly: I'm blocked on the "working example" part. CodeRabbit is a hosted PR-review service — its workflow is: you install its GitHub App (or GitLab integration) on a real repo via its web dashboard, then it reviews pull requests automatically. There's no standalone SDK or public REST API I can call with just an API key; activation requires an account, a paid/trial plan, and a connected remote repo, none of which exist here. I can't trigger or observe a review without that access, so I'll stop short of pretending to build a live integration. No local stack or long-running commands are involved — it's purely a credentials/access blocker.
What I can do lightly is drop the one repo-side artifact that is a genuine developer workflow — the config file CodeRabbit reads when installed:
File edits
1 file
Written.coderabbit.yamlView changes
/sandbox/repo/.coderabbit.yaml
Written content · previous contents were not recorded
# CodeRabbit configuration — takes effect once the CodeRabbit GitHub App# (or GitLab integration) is installed on this repo.# Docs: https://docs.coderabbit.ai/reference/configurationlanguage: en-UStone_instructions: "Be concise and pragmatic; prefer actionable suggestions."reviews:profile: chill # "chill" (less noisy) or "assertive"auto_review:enabled: truedrafts: falsereview_status: truechat:auto_reply: truepath_instructions:- path: "**/*.py"instructions: "Follow PEP 8; require type hints on public functions."
Source: write · chatcmpl-tool-8c0a4a2aa70e49c8
Recorded tool call
{
"type": "tool_call",
"seq": 12,
"ts": "2026-10-06T16:37:07.165Z",
"name": "write",
"turn": 1,
"input": {
"path": "/sandbox/repo/.coderabbit.yaml",
"content": "# CodeRabbit configuration — takes effect once the CodeRabbit GitHub App\n# (or GitLab integration) is installed on this repo.\n# Docs: https://docs.coderabbit.ai/reference/configuration\n\nlanguage: en-US\ntone_instructions: \"Be concise and pragmatic; prefer actionable suggestions.\"\n\nreviews:\n profile: chill # \"chill\" (less noisy) or \"assertive\"\n auto_review:\n enabled: true\n drafts: false\n review_status: true\n\nchat:\n auto_reply: true\n\npath_instructions:\n - path: \"**/*.py\"\n instructions: \"Follow PEP 8; require type hints on public functions.\"\n"
},
"run_id": "cmuwwigqd01fk0ivpotoe66fu",
"tool_id": "chatcmpl-tool-8c0a4a2aa70e49c8",
"tool_kind": "file_edit"
}Recorded tool result
{
"type": "tool_result",
"seq": 13,
"ts": "2026-10-06T16:37:07.165Z",
"turn": 1,
"output": {
"content": [
{
"text": "Successfully wrote 567 bytes to /sandbox/repo/.coderabbit.yaml",
"type": "text"
}
]
},
"run_id": "cmuwwigqd01fk0ivpotoe66fu",
"tool_id": "chatcmpl-tool-8c0a4a2aa70e49c8",
"is_error": false,
"tool_kind": "file_edit"
}›Thinking…
Next, I'll briefly answer about pricing and ease of management.
Created /sandbox/repo/.coderabbit.yaml — that's the extent of what's actionable here.
How pricing works (check coderabbit.ai/pricing for current numbers, as these change):
- Free / Lite — free for open-source projects and personal use; unlimited reviews on public repos with some feature limits.
- Pro — paid per seat per month (~$12/seat for individual devs, ~$24/seat for teams), adds full features (assertive profile, custom instructions, analytics, etc.) for private repos.
- Enterprise — custom pricing, adds SSO/SAML, self-hosted options, priority support.
- Free Pro trial on install; billing is seat-based on your GitHub/GitLab org.
Will it be easy for me to manage? Mixed:
- ✅ The repo-side surface is easy:
.coderabbit.yaml, path instructions, and per-PR control via chat commands (@coderabbitai review,@coderabbitai resolve) are all plain files/comments I can edit through git. - ❌ The activation and interaction side is not mine to reach: installing the app, managing the subscription, and reading review comments require the web dashboard or a GitHub/GitLab API token. With a GitHub token provided, I could interact with its PR comments via the GitHub API and manage it reasonably well; without one, it's mostly hands-off for me.
Bottom line: blocked on a live example (no account, no app install, no remote repo/credentials). If you connect CodeRabbit to a real repo and give me a GitHub token, I can take it from there.
AX Fundamentals
23 checks. Open an area for details.
This check is a quick preview. Test more tasks and models with Gauge Agents.
Test your product with Gauge Agents