Your email was sent to websitesClaude Code · 91 times · last Nov 7
API keys are sitting in your agent logsClaude Code · 3 times · today
History on main was overwrittenGit · 4 times · last Nov 7
Medium 2
Commits were made without your repo's checksGit · once · last Nov 13
A form was submitted under your nameClaude Code · once · last Nov 6
Fixed and hidden 1
Finding
API keys are sitting in your agent logs
High · Claude Code · 3 times · today
2 keys that look real were typed into commands, so they are stored in plain text in 3 log files.
Remove keys from the logs…Removed 2 keys from 3 logs.
Examples 1
S
export STRIPE_KEY=‹key› && node refund.jsshop · 9:12 AM
Protection
Agents 3/3Rules Sandbox OnGit 4/4Keys 0Network 2
Only one of your agents is told 1 of your rules.Claude Code and Codex are told the same rules.
Rules from CLAUDE.md and AGENTS.md that the others never read.Rules from CLAUDE.md and AGENTS.md, counted below.
Share Rules…
3Never delete
4Never open
2Never run
0Told to one agent only
Who is told what
RuleClaude CodeCodexWritten in
never run terraform destroy~/Sites/infra/CLAUDE.md
never open ~/.vault~/.claude/CLAUDE.md
never delete ~/backups~/.claude/CLAUDE.md
never run npm publish~/Sites/shop/CLAUDE.md
never open ~/.ssh~/.claude/CLAUDE.md
Protection is on
3 blocked today·checked just now
Notices
1 of your rules reaches only Claude CodeShare it with the other agent
Needs you 1Review all
Claude Code: open ~/.vaultContinue
Running now 2
Codex in shop1mStop
Gemini CLI in citybot4mStop
Alerts 3 newMark all as seen
Claude Code: open ~/.vault9:41 AM
Blocked: your email to a website9:41 AM
Blocked: a read of ~/.aws/credentials9:20 AM
Show History…
Open Excellent⌘O
Settings…⌘,
Quit Excellent⌘Q
Remove keys from the logs?
Overwrite keys in known formats in the logs Excellent can edit. Logs used by a running session stay unchanged until that session ends.
Understand how your agent works. Catch mistakes and give your agent clearer rules for next time
Overview
OverviewChecked 1,248 sessions · Claude Code, Codex · just now
Sessions checked
1,248
Tool calls
61,904
Need you
5
Fixed
2
Sessions
Update the pricing pageClaude Code · docs-site · 2 minutes ago
Review the sync changeCodex · api-server · 6 minutes ago
Excellent reads the logs your agents already keep
Your agents keep a diary of what they did, helpers included. Press Scan Now, and Excellent reads them, on your Mac. A first scan takes seconds.
Findings
Findings
Your email was sent to websitesHigh · 34 times · last Nov 17
API keys are sitting in your agent logsHigh · 2 times · last Nov 14
History on main was overwrittenHigh · 1 time · last Nov 16
Your email was typed into web formsMedium · 2 times · last Nov 15
Uncover failures, security issues and vulnerabilities
Your email sent to websites or typed into forms. Keys sitting in logs. Secrets files read. History overwritten. Checks skipped. Each one says which agent and where.
What you can do
Finding
Your email was sent to websites
High34 times · Oct 30 to Nov 17
34 commands put your email in requests to 5 websites. 9 came after Nov 12.
What you can do
Block this from now onAdds a check that stops your agents sending your email to a website.
Remove it from 1 scriptAsks your agent to replace your email with your website in ~/Sites/portfolio-site/fetch_refs.py, then checks the file.
Adds a check that stops your agents sending your email to a website.
Fix it with a click and monitor what happens next
One click fixes some issues. Your own agent cleans up what it wrote, and Excellent checks the files again. Keys are cleared from logs. Broken hooks are repaired. For the rest, with Pro, a button turns on a check, so Excellent watches for it next time. Excellent backs up a settings file before it changes it. Clearing keys from logs can't be undone.
never delete ~/modelsEverywhere · ~/.claude/CLAUDE.mdShared with an agentNow shared with another agent
never open ~/financeEverywhere · ~/.claude/CLAUDE.mdShared with an agentNow shared with another agent
never run `git add -A`Only in ~/Sites/billing-app · ~/Sites/billing-app/CLAUDE.mdShared with an agentNow shared with another agent
Share…
Give your agents the same rules
Write your rules once, in CLAUDE.md or AGENTS.md. With Pro, Excellent's live check holds each agent it runs in to them, and asks you before one breaks a rule. A scan shows which agent broke one.
$ cat ~/finance/taxes-2025.pdfcat: /Users/jordan/finance/taxes-2025.pdf: Operation not permitted
$ python3 -c "open('/Users/jordan/finance/taxes-2025.pdf').read()"PermissionError: [Errno 1] Operation not permitted
Let Excellent enforce them
Pro writes your file and folder rules into the macOS sandbox of the agents that support it. Then the Mac itself keeps them out of the folders you closed, however a command is worded. Undo takes it out.
Privacy
Goes onlineNo account needed · Usage stats only if you turn them on
Sent to Excellent
Your logs never come to us.
Signed in for Pro
A check of your plan, once a day
Check GitHub and GitLab
When you start it
Protect Main
When you start it
Servers preview
Every 5 minutes, once on
Explain and agent fixes
Your own agent, to its provider
Nothing about you is sent to us
Unless you sign in for Pro, share usage stats or pair it with our team, nothing about you or your agents is sent to us.
Needs Python 3.9 or later. Reads only. Makes no network calls. Writes one file: ~/excellent-free-scan/report.html. Read the source
Check your agents’ work
Free · Manual
Download Excellent for Mac
Free to scan. No account needed.
Open it and press Scan Now to see what your agents did. With Pro, it also checks risky actions before they run, and it lists what it will change in your agents' settings before it changes anything.
It reads the logs your agents keep on your Mac and shows what they did: your email sent to websites, keys in logs, secrets files read, history overwritten, rules that never reached the agent. It can fix some of it: clear keys from logs, take your email out of scripts, share rules and repair broken hooks. With Pro, a button turns on a check for next time, and Excellent checks each action before it runs. Excellent for Mac today; a version for Linux servers comes next, then Windows.
Does anything leave my Mac?
Unless you sign in for Pro, share usage stats or pair it with our team, nothing about you or your agents is sent to us. Excellent reads your agents' logs on your Mac and keeps what it finds there. The free app needs no account. No analytics unless you turn on usage stats, which send counts, never your logs. It checks for updates about once a day, which you can turn off. While you're signed in for Pro, it asks excellent.so about once a day whether your account has Pro, sending only its sign-in token. A few things leave when you start them, such as Explain, agent fixes, Check GitHub and GitLab, Protect Main and the Servers preview. Explain hands a masked finding to your own agent, on your plan. An agent fix gives it the file's path, and it reads and edits that file itself. Replace this key opens the provider's key page in your browser. You make the new key there. The check's replies, which can quote your rules, reach your agent's provider.
Which agents does it work with?
It works with Claude Code, Codex, Cursor, GitHub Copilot CLI, Gemini CLI, Grok Build, Qwen Code, Factory Droid, Devin and other AI agents. The agents page says what Excellent does with each one.
What does it change on my Mac?
A scan changes nothing. Protection, with Pro, changes nothing until you say so: before you turn it on, it lists what it will change. It adds one check to each agent's settings, backing up each file first. Where an agent asks for it, it also tells that agent to trust it, and it starts at login. Turn it off and it takes its check back out of each agent's settings, and changes nothing else. Agent trust, shared rules, a sandbox and commit watching stay until you undo each. It never opens your Documents, Desktop, Downloads or iCloud folders.
Is it free?
Scanning is free. Free shows what your agents did: scan any time, read every finding and its history, and get every app update, with no account needed. Pro protects: it blocks or asks before risky actions, turns your rules into the Mac's own sandbox, scans in the background and alerts you. Pro is $6 a month or $60 a year. Pro unlocks the moment you pay. In January, Pro adds your other Macs and servers in one view, and alerts on your phone or in Slack.
Can an agent get around it?
Pro's live check reads what a command says. An agent set on getting around it sometimes can. For the file and folder rules that matter most, Pro can turn them into the macOS sandbox of the agents that support it. The sandbox doesn't care how a command is worded.
Why did you build it?
Helper agents put my email address into more than 150 web requests, and I found out by accident. I wanted to see what my agents did, so I built this.