launch-codes
Dangerous Bash commands need launch codes: red alert, siren and a LAUNCH key.
Illustration. Actual results depend on the mod, terminal and environment.
Install
/plugin marketplace add OneWave-AI/claude-code-mods
/plugin install launch-codes@claude-code-mods
/reload-pluginsRun these commands one at a time in a Claude Code session.
claude plugin marketplace add OneWave-AI/claude-code-mods
claude plugin install launch-codes@claude-code-modsThen run /reload-plugins in your session, or start a new session.
Please install the Claude Code mod "launch-codes" (https://github.com/OneWave-AI/claude-code-mods). Before installing, read the Security and Requirements sections of its README and tell me what it reads and writes, which commands it runs, whether it uses the network, and which Claude Code version, OS and terminal it needs. If everything looks fine, run claude plugin marketplace add OneWave-AI/claude-code-mods and claude plugin install launch-codes@claude-code-mods, then remind me to run /reload-plugins in my session.Send this to Claude Code so it reads the author’s instructions before installing.
git clone https://github.com/OneWave-AI/claude-code-mods
cd claude-code-mods
claude --plugin-dir ./launch-codesClone the repository and try it in a local session with --plugin-dir.
What it accesses
Based on the author’s documentation, checked when listed; use claude plugin validate to check for yourself.
Reads
YesChecks the command text of Bash tool calls.
Writes files
NoneRuns commands
NoneNetwork
NoneIntroduction
launch-codes is a safety mod from OneWave AI. It checks Bash commands before they run; a risky one is held, a red-alert pane opens with a siren and a one-time code, and only typing the code to arm and pressing LAUNCH runs it. Anything else is denied and Claude is told why.
- Catches
rm -rf,git push --force,git reset --hard,DROP/TRUNCATEthrough a live SQL client,supabase db reset,vercel --prod,chmod -R 777, and downloaded scripts piped into a shell /launch-codes test <command>shows the verdict without running anything
The authors say it is strict and blocked one of their own cleanup commands. It checks Bash only, so it is a speed bump rather than a security boundary and will not stop an agent that writes a script and runs it another way; keep normal permissions and sandboxing on. It makes no network calls, runs no processes, touches no files and calls no model. Requires Claude Code 2.1.287+. Not run-tested by this site; the preview is illustrative.
Who it is for
- People who want one more check before destructive commands
- People working close to production databases or main branches
Requirements
- Claude Code ≥ 2.1.287
- Checks Bash only and is not a security boundary; keep permissions and sandboxing on.