Safety & Privacy · Claude Code mod budget-guard
Stops a session before it runs past a budget: warns near a cost or plan-window limit, then refuses tool calls and asks before the next prompt.
Swaps secrets, email addresses and IPs for stable placeholders before the model reads them, and restores them on the way into a tool call.
Safety & Privacy mod for Claude Code by ray-amjad · built on function hooks · runs inside Claude Code
Step 0, once: enable function hooks with export CLAUDE_CODE_ENABLE_FUNCTION_HOOKS=1 (bash/zsh), $env:CLAUDE_CODE_ENABLE_FUNCTION_HOOKS="1" (PowerShell), or put it under env in ~/.claude/settings.json. Needs Claude Code 2.1.269+. Guide
claude plugin marketplace add ray-amjad/awesome-claude-code-function-hooks claude plugin install secret-redactor@awesome-claude-code-function-hooks
[](https://claudemods.ai/mods/secret-redactor) The model sees <SECRET_1>, the tool gets the real value. One of the first two function-hook plugins published, from before the "Mods" name.
New to mods? Start here. claudemods.ai is an independent catalog, not affiliated with Anthropic. Listings are contributed by the community and link to third-party code. Read the source before installing; mods run inside Claude Code with the permissions of your session. Something wrong with this listing? Tell us.
Safety & Privacy · Claude Code mod Stops a session before it runs past a budget: warns near a cost or plan-window limit, then refuses tool calls and asks before the next prompt.
Safety & Privacy · Claude Code mod Client-side redaction of API keys and sensitive identifiers from Read, Bash and Grep output before it reaches the model.
Safety & Privacy · Claude Code mod PII de-identification as a mod: placeholders go to the model, real values are restored before Edit, Write and Bash, exfiltrating tools are refused.