A small script that catches invisible characters in plugin names before you install
Control characters and zero-width characters hiding in plugin or skill names are invisible in a terminal listing. I built six deliberately confusing names, compared what ls actually shows, and wrote a short check you can run before installing anything — plus an honest note on what that check misses.
My read-deny rules never reached inside the folder I had allowed
An audit of how far permissions.deny Read rules actually reach in a working tree where secrets and source code live side by side, with a script you can run today.
Take Inventory of What You Allowed with Don't Ask Again in Claude Code
After clicking Don't ask again a few dozen times, can you still explain what you approved? Here is a script that counts your current rules, plus a way to rebuild permissions from the smallest possible set.
Passing the Request, Not the Secret — Where Sandbox Credential Masking Works and Where Substitution Breaks
Claude Code's sandbox credential masking lets processes read sentinel values while a proxy swaps in the real secret at send time. I rebuilt it as a minimal proxy and measured which auth schemes survive the swap, which break, and what happens when the secret rides in the body.
Tightening Filesystem Isolation Separately from the Network — Collect the Paths, Then Squeeze the Write Surface
Claude Code v2.1.216 lets you control filesystem isolation independently from network isolation. Before tightening anything, I traced what a real job actually touches, split reads from writes, and measured how stable the path set is across repeated runs. The numbers changed how I wrote the allowlist.
You Added a Second Working Root — Now Ask How Far Your Ignore Rules and Deny Patterns Actually Reach
Adding a working root mid-session does not carry your ignore rules or deny patterns with it. Measured evidence from git check-ignore and a three-way matcher comparison, plus the scope-delta audit script I now run from the DirectoryAdded hook.
Locking down Claude Code sandbox egress with strictAllowlist
Tightening automation egress with strictAllowlist in Claude Code v2.1.219, plus measured failure timings that tell a policy deny from DNS and real outages.
A Committed Symlink That Points Outside the Worktree — Auditing Repos Before You Let AI Spin Up Parallel Trees
Claude Code 2.1.212 fixed a bug where a committed symlink under .claude/worktrees could be followed during worktree creation and write outside the repo. The patch closes the following side. Here is an audit script for the committed side, plus a quarantine workflow.
I Believed Plan Mode Only Read — Replacing That Belief With Machinery
Claude Code 2.1.212 fixed a bug where plan mode ran file-modifying Bash commands without the permission prompt or the SDK canUseTool callback. Here is what could happen while that assumption was broken, how to verify your own setup, and how to stop leaning on a mode name for safety.
The String I Approved Wasn't the String I Read — Testing a Relayed Permission Prompt with Deceptive Characters
I pushed bidi overrides and zero-width characters through my own approval relay. NFKC normalization caught 0%. Here is why, and the implementation that catches 100% with zero false positives.
The Night a Connector Grew Write Tools, Your Unattended Job Could Quietly Send Something
The moment a connector update quietly adds write tools, a job you run unattended can suddenly send email or delete files. Here is how to build a gate that snapshots a connector's tool surface, diffs it, and stops unapproved writes before they run — drawn from indie development.
When You Give an API Key an Expiration Date, Expiry Becomes a Plan Instead of an Accident
The Console now lets you set expiration dates on API keys. Here is how to fold planned expiry into unattended operations — with overlapping dual keys and a local expiry ledger — so your nightly jobs never go dark.