Claude Code
Today
Explore
Practice
Library
Exam
Certificate
Search
⌘K
ខ្មែរ
◐
Sign in
Today
Explore
Practice
Library
Profile
The Permission Model
Today
3. Permissions and Safety · 1 of 4
Mark read
Prev
Next
Orient
Three tiers, two persistence behaviors
Find where Monday's approval was written
Predict
Audit the approvals already in your repo
Close
Practice
Related
The Permission Model
Intermediate
☆
📑
Enforced by Claude Code, not by the model — and that distinction is the whole safety story
Orient
Three tiers, two persistence behaviors
Find where Monday's approval was written
Predict
Audit the approvals already in your repo
Close
Practice
Related
Practice activities and scenarios are optional. Mark the lesson complete after reading, and return to practice whenever it helps.
Restoring your saved work…
07
Optional practice
🧠 Optional scenario practice
Q1
You approve a Bash command with "Yes, don't ask again," and separately approve a file edit the same way. What persists?
A
Both, in .claude/settings.json
B
The Bash rule persists to .claude/settings.local.json at the repo root; the file-edit approval lasts only until the session ends
C
Neither — both are session-scoped
D
Both, but only in the directory you started from
Q2
With an allow rule of
Bash(safe-cmd *)
, will
safe-cmd && rm -rf build
run without prompting?
A
Yes — the command starts with safe-cmd
B
No — Claude Code splits compound commands and each subcommand must match a rule independently
C
Yes, but only if rm is in the read-only command set
D
Only in acceptEdits mode
08
Up Next
Advanced
Writing Rules That Hold
Advanced
Modes and Sandboxing
Advanced
Untrusted Content and Injection
📝 Notes
Mark lesson complete
Next: Writing Rules That Hold →