[Resource]: claude-code-guardian - Permission system for CC, define rules for what it can execute, read or write

[jfpedroza]

Display Name

Claude Code Guardian

Category

Hooks

Sub-Category

None

Primary Link

https://github.com/jfpedroza/claude-code-guardian

Secondary Link

No response

Author Name

jfpedroza

Author Link

https://github.com/jfpedroza

License

MIT

Other License

No response

Description

Validation and permission system for Claude Code focused on controlling what Claude Code can execute, read or write. Allowing users to define a set of rules to evaluate.

Additional Comments

I wanted to have more control on what Claude could execute by without asking. Allowing certain commands without permitting dangerous versions of them. E.g. allow git push but not git push --force.

Submission Checklist

• [x] I have checked that this resource hasn't already been submitted
• [x] My resource provides genuine value to Claude Code users, and any risks are clearly stated
• [x] All provided links are working and publicly accessible
• [x] I am submitting only ONE resource in this issue
• [x] I understand that low-quality or duplicate submissions may be rejected

[github-actions[bot]]

🤖 Validation Results

✅ All validation checks passed!

Your submission is ready for review by a maintainer.

Validated Data:

{
  "display_name": "Claude Code Guardian",
  "category": "Hooks",
  "primary_link": "https://github.com/jfpedroza/claude-code-guardian",
  "secondary_link": "",
  "author_name": "jfpedroza",
  "author_link": "https://github.com/jfpedroza",
  "license": "MIT",
  "description": "Validation and permission system for Claude Code focused on controlling what Claude Code can execute, read or write. Allowing users to define a set of rules to evaluate.",
  "subcategory": "",
  "active": "TRUE",
  "last_checked": "2025-08-16:08-12-02"
}

---

_{This comment is automatically updated when you edit the issue.}

[hesreallyhim]

@jfpedroza i've been figuring out a evidence/data-driven way to evaluate tools like this. so, e.g., something like this which makes reviewing it kind of tricky is: it describes a specific permissions/boundary system, right? so in order to know if it's effective, i have to try to do something which might otherwise be unsafe if i didn't have claude-code-guardian set up (git push --force or something). well, ok, that requires a bit of trust in the quality of the tool already, because if it happens not to work, oops I messed up my project - but what would be very really helpful is a simple, clear statement of a scenario/context where i could use this tool, describe a prompt that i can give to claude code, and explain what i should expect to see when using claude-code-guardian vs not. Sounds good? /request-changes see comment - request for clear demo scenario

[github-actions[bot]]

🔄 Changes Requested by @hesreallyhim

see comment - request for clear demo scenario

Please edit your issue to address these points. The validation will run again automatically after you make changes.

[jfpedroza]

Hi. I agree that testing the tool with a dangerous command is a bad idea. What I do when testing changes is create a rule like "deny writing to a .deny file" and then ask Claude to "write world into hello.deny" and see if it works. I will update the issue with the scenario later.