
Bitwarden's review skills check the agents, commands, hooks, settings and CLAUDE.md in your Claude Code setup for security and quality.
by bitwarden· SecurityUnclaimed · Yours? Claim it
About
Bitwarden's own Claude Code plugin for reviewing Claude configuration. It checks the agents, slash commands, hooks, settings and CLAUDE.md files in a repository or a plugin for security, structure and prompt quality. It reports only what a change introduced, one finding per issue, pinned to its file and line. It's the same review Bitwarden's validate-ai GitHub Action runs on its own pull requests. What it checks - Agents: tool access against least privilege, frontmatter, descriptions and triggers, system prompts and model choice - Slash commands and prompts: purpose, session context, skill references and parameters - Settings and hooks: permission scoping, auto-approved commands, and every hook command read as code that will run - CLAUDE.md: clarity, specificity and length, since it's re-read every turn - Security first, every time: a committed settings.local.json, hardcoded secrets and overly broad permissions How you use it - Ask in your own words, such as "Review my new agent in .claude/agents/", or run /claude-config-validator:reviewing-claude-config - /validate-ai-local reviews what you changed on your branch, plus uncommitted work, and writes a report - /validate-ai reviews a pull request and posts a summary comment to it through the GitHub CLI - Findings are ranked CRITICAL, IMPORTANT, SUGGESTED or OPTIONAL, each with a fix and why it matters How to install it In Claude Code, run /plugin marketplace add bitwarden/ai-plugins, then /plugin install claude-config-validator@bitwarden-marketplace. Good to know - It needs Claude Code 2.1.210 or later. - Skill and plugin-manifest checks come from Anthropic's plugin-dev plugin. Without it, the commands report those sections as skipped. - It also includes a security-scan script you can run yourself on a .claude folder.
Plugins people install alongside Claude Config Validator.
4 images
Once it’s added
“Review my new agent configuration in .claude/agents/code-analyzer.md”
“Review the settings and hook changes in this branch”
“Security audit all Claude configuration files in this project”
“Review my plugin in plugins/my-plugin/ for marketplace readiness”
“Validate the Claude config I changed on this branch before I push”
“Validate the Claude changes in pull request 196”
Before you add it
Reads only
Your agents, commands, hooks, settings and CLAUDE.md files · Your branch's changes, or a pull request's diff
Makes changes · your agent can ask first
A validation report in the plugin's data folder (/validate-ai-local) · A summary comment on the pull request (/validate-ai), which you approve first
From people who use it
Step by step
Nothing new reaches your agent until a person has looked at it.
Scanned on every change, and again when our rules improve.
Runs on your machine.
Runs on your machine.
/plugin marketplace add bitwarden/ai-plugins /plugin install claude-config-validator@bitwarden-marketplace
Scanned and passed · checked 6 h ago
No reviews yet. Used Claude Config Validator? Say how it went.
Write a reviewPlugins people install alongside Claude Config Validator.
Anthropic's security review for Claude Code: flags injection, XSS, secrets and more as Claude edits, finishes a turn and commits.
Equips your coding agent with SAST, dependency scanning, a phased hardening workflow and STRIDE threat-modeling skills.
Alireza Rezvani's Claude Code hook that stops Claude before it writes common security anti-patterns, such as eval, pickle or SQL f-strings.
Microsoft's skills for Azure Data Explorer: build Kusto graphs, compose IRQL threat hunts and turn results into visual graphs.
Nobody has shared how they use Claude Config Validator yet.
Security scan
Passed
Safe to install in every agent it supports. Claude plugins.