Claude Code Safety and Permission Mods

Clawd's pixel world above your prompt in the Claude desktop app, in five scenes (open space among them, where everyone floats) that follow your local…

Has Cloudflare's Clef decision model judge each Bash, Write, Edit and WebFetch call (risk, kind of risk, whether it is read-only, whether it fits…

Mask client names, personal data and secrets before anything leaves this machine; keep the list private and leave no trace of what was swapped

Refuses destructive Bash commands (rm -rf /, force-push to main, DROP TABLE, curl | sh, ...) before Claude runs them, and tells Claude why

Claude Code status line plugin: context %, session cost, budget, context and rate-limit alerts, /cockpit dashboard with rate limits, burn rate and…

Cost band above the prompt, a live team pane for subagents, and a blast-radius guard for wide commands

A rubber duck in your Claude Code session: a quiet second opinion on each turn, a guard for risky commands, reminders, and lessons about your codebase

Holds risky tool calls (production domains, live Stripe keys, protected-branch pushes, deploys, destructive commands) for your OK, replays each…

Guided modernization for any legacy codebase: start with /modernize, get an assessment, an interactive map, the business rules mined from the code…

Guided modernization for any legacy codebase: start with /modernize, get an assessment, an interactive map, the business rules mined from the code…

Guided modernization for any legacy codebase: start with /modernize, get an assessment, an interactive map, the business rules mined from the code…

Pi-style codemode: run JavaScript in a QuickJS sandbox that calls Claude Code's tools

Asks what to do before Claude Code re-sends a large conversation whose prompt cache has expired: on a cold resume, and on the first prompt after an…

Blocks git commit and git push when the change contains secrets, private keys, .env files, node_modules or giant files, before they leave your…

Context runway gauge: warns before your context line with Compact, Hand off and Snooze

Live context meter above the prompt; asks Yes/No to save a handoff and clear when the window fills, and feeds that handoff into the next session

A permanent, slim context-usage bar above the prompt with a chomping Pac-Man and a 10px bar; green under 20%, amber 20-30%, red past 30%, munching…

Two rules no settings hook enforces, refused at the tool call: wiki pages and their generated ADR mirrors are written only through the Cortex wiki…

Selective continuity protection with session-aware budgets, exact user evidence, and verified handoffs

Holds destructive shell commands (rm -r, force pushes, hard resets, DROP TABLE) for your confirmation and says what they would hit

Compacta sozinho ao fim do turno quando o payload da conversa passa de 16 MB, inclusive em sessões headless (-p / SDK)

Asks you before a command that can destroy work, such as rm -rf, git reset --hard or git push --force

Before the agent runs a Bash command, blocks the ones that can wreck the machine or the user's data beyond the working repository: sudo, deletes of…

Deckhand — the multi-AI toolbar for Claude Code

De-identifies personal data in tabular files (CSV, TSV, XLSX, JSON) before anything reaches the model

A pane that lists the files a delete command would remove, with a keep/delete toggle per file and Confirm or Refuse buttons

Deleting outside the project folder requires the user's approval (Bash, PowerShell, cmd, subagents, MCP)

Context usage, branch and tool count in a band; tokens per request charted in a pane; a note tool for Claude and a check before dangerous commands

Chimes when a long turn finishes and when a permission prompt is waiting; /chime to toggle

Claude Code mod: blocks Claude from reading .env secret files and counts blocked attempts in the status line

Explains each Claude Code permission request in plain English (WHAT, WHY, RISK) in a side pane

Side panel for Claude Code: pin rules the model keeps through /compact, flip live guards (read-only, no push, no edits on main, fence, secret…

Live panel and wait-for-returns guard for the self-staffing-focus-group skill: roster, gates G0-G5, findings; reviewers and verifiers run in the…

Grammar fixes for prompts you type or dictate: optional auto-fix on send (LanguageTool rules or a light Haiku fix), Haiku drafts that know your…

Blocks risky shell commands (phone deletes, printing secrets, self-killing pkill) and strips secrets pasted into messages

Tapa las llaves y contrasenas antes de que Claude las vea, y le prohibe imprimir archivos de secretos

Pide confirmación con botones antes de que Claude ejecute comandos peligrosos (rm -rf, git push --force, git reset --hard)

Muestra cuánto le queda al caché, el contexto y los límites, con botón de traspaso de sesión

Blocks the Bash commands your own rules file names, and shows every block in /blocked

A human signs off before anything that can't be undone: publishing, killing protected processes, editing secrets or stale copies

Pattern-based checks that deny .env reads, git force pushes and rm -rf on / or your home folder before Claude Code runs them

Turn CLAUDE.md rules into hard blocks: protected paths, no new report files, banned commands, and rules re-pinned every N prompts and after…

Enforces your standing rules for Claude: no push to main, no force push, no surprise deletes or new dependencies

Jev answers clear permission prompts, flags prompt injection in fetched content, checks done claims, nudges large or vague requests, and routes…

Denies a delegated agent lane any write to this repo's Claude settings files

Idle-session guard for Claude Code's 1-hour prompt cache (macOS): about 50 idle minutes after the last turn of a large session it notifies you…

Keep your API keys and tokens out of your Claude Code conversations: tokens pasted as keyfob: lines are stored before the model sees them, Claude…

Masks credentials in tool output (URL passwords, *_TOKEN / *_SECRET / *_PASSWORD values, API keys, bearer tokens, private keys) before Claude or the…