Runtime Protection for Claude Code Agents
Noma Security provides active runtime protection for Claude Code by sitting between your AI agents and their intended actions. This plugin enables you to evaluate, allow, or block high-risk activities in real-time.
For more details, visit noma.security.
With Claude Code Hooks enabled, Noma acts as a security gatekeeper for the following high-risk agent actions:
.env files, SSH keys) from being indexed or sent to the LLMAdd the Noma marketplace to your Claude Code instance:
claude plugin marketplace add https://github.com/Noma-Security/claude-marketplace
Install the specific Noma guardrails hook:
claude plugin install guardrails@noma-marketplace
To connect Claude Code to your Noma instance, you need to configure the NOMA_API_KEY.
If your organization manages Claude Code usage via a centralized settings.json, your administrator can push these configurations directly:
settings.json to include the Noma environment variablesFor individual setups, instead of exporting variables in your shell profile, Claude Code reads configurations from a local JSON file. This mirrors the structure used in the Managed Settings method.
To configure your local environment, add your Noma credentials to the following path: ~/.claude/settings.json Example settings.json structure:
{
"NOMA_API_KEY": "your-secret-api-key"
}
Once the plugin and settings are in place, authorize the managed settings within Claude Code:
If you installed the plugin during an active session, refresh the state:
/reload-plugins
To confirm that Noma is actively protecting your session:
Application ID -> Claude-Code to see real-time allow/block eventsLook for Debug mode indicators and status bar labels to confirm protection is active.
claude plugin list to ensure guardrails@noma-marketplace is listed and active/reload-plugins after making changes to your plugin configuration. Restart Claude after every environment variable changesettings.json follows the correct Claude Code schema. Invalid syntax will cause the CLI to ignore managed settingsclaude auth status401 Unauthorized or 403 Forbidden errors related to NomaNote: Claude Code Hooks is currently in Beta status. Beta status means Noma is actively researching, iterating, and developing this feature. Based on feedback, market innovation, and technical and commercial viability, Noma may decide to suspend further work on this feature. To gain early access to a beta feature initiative, contact your Noma Technical Account Manager.
For support and access to beta features, contact your Noma Technical Account Manager.
Noma Security handles security for AI, providing comprehensive protection for AI-powered development tools and workflows.
Matches all tools
Hooks run on every tool call, not just specific ones
Own this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimOwn this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimBased on adoption, maintenance, documentation, and repository signals. Not a security audit or endorsement.
npx claudepluginhub noma-security/claude-marketplace --plugin guardrailsAchieve flow state safely with Claude Code. Auto-approves routine work, gates risky actions, hard-blocks dangerous patterns. Dual enforcement (skill + hooks), token cap for cost governance, full audit trail. Zero dependencies.
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.
Command Line Agent Safety Harness. All interactions with clash policy should go through this plugin
Runtime security for AI agents. Blocks destructive actions before execution, routes high-risk operations through human approval, and maintains an immutable audit trail. Covers OWASP MCP Top 10, ASI Top 10, and Agentic Skills Top 10.
GoPlus AgentGuard — AI agent security guard. Blocks dangerous commands, prevents data leaks, protects secrets. 20 detection rules, runtime action evaluation, trust registry.
Protect secrets and credentials from Claude Code. Blocks writes to .env files, detects API keys in shell commands, prevents hardcoded tokens, and guards service account JSON files.