By kirich1409
Intercept bash tool calls to scan files for secrets and PII before sending to AI servers, blocking sensitive data leaks during conversations. Uses PreToolUse hook with Node.js script to analyze and halt risky file reads.
npx claudepluginhub kirich1409/krozov-ai-tools --plugin sensitive-guardClaude Code plugin marketplace by Kirill Rozov.
Add the marketplace to Claude Code:
/plugin marketplace add kirich1409/krozov-ai-tools
Install a plugin:
/plugin install maven-mcp@krozov-ai-tools
/plugin install sensitive-guard@krozov-ai-tools
/plugin install developer-workflow@krozov-ai-tools
/plugin install extend@krozov-ai-tools
Maven dependency intelligence for JVM projects. Auto-registers an MCP server that provides tools for version lookup, dependency auditing, vulnerability checking, and changelog tracking across Maven Central, Google Maven, and custom repositories.
Features:
Skills: /check-deps, /latest-version, /dependency-changes
See plugins/maven-mcp/ for full documentation.
Prevents sensitive data (secrets, PII) from reaching AI servers. Scans files via a PreToolUse hook before they are read into conversation.
Features:
See plugins/sensitive-guard/ for full documentation.
Developer workflow skills and expert agents for the full development cycle — from task implementation to QA testing to PR merge.
Skills:
/prepare-for-pr — quality loop (build → simplify → self-review → lint/tests) before creating a PR/create-pr — create a draft or ready PR with auto-generated title, description, labels, and reviewer suggestions/pr-drive-to-merge — drive an existing PR/MR to merge: monitor CI/CD, triage reviewer comments, loop until merged/code-migration — safe in-place or parallel migration of any technology in Gradle/Android/Kotlin projects/kmp-migration — full Kotlin Multiplatform migration for Android modules/migrate-to-compose — migrate View-based Android UI (Activity, Fragment, custom View) to Jetpack Compose/generate-test-plan — generate structured, prioritized test plan from spec or code/test-feature — verify a feature against its specification on a live app/exploratory-test — undirected bug hunting and QA exploration on a running app/plan-review — multi-agent review of implementation plans using PoLL consensus protocol/implement-task — full development cycle: worktree → TDD → implementation → quality loop → draft PR (explicit invocation only)Agents (10): architecture-expert, build-engineer, business-analyst, compose-developer, devops-expert, kotlin-engineer, manual-tester, performance-expert, security-expert, ux-expert
See plugins/developer-workflow/ for full documentation.
Extend Claude Code built-in features with review and optimization tools.
Skills:
/agent-reviewer — audit and improve Claude Code agent files: frontmatter, system prompt quality, tool selection, trigger accuracySee plugins/extend/ for full documentation.
MIT
Blocks secrets and PII before they reach the Anthropic API
Scan for data privacy issues
Security reminder hook that warns about potential security issues when editing files, including command injection, XSS, and unsafe code patterns
Safety for Agents - Agent Detection & Response (ADR) for Claude Code
5 essential safety hooks for Claude Code. Blocks rm -rf, force-push, hard-reset, .env overwrites, and package publish. The minimum viable safety net from 800+ hours of autonomous operation.
Executes bash commands
Hook triggers when Bash tool is used
Share bugs, ideas, or general feedback.
A secure runtime for Claude Code. Intercepts every tool call with policy-based allow/block/ask decisions, evasion detection, path fencing, file snapshots, and audit logging.
Own this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge.
Sign in to claimOwn this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge.
Sign in to claim