# Infisical MCP for AI Agents AI Agent Connect > Infisical MCP. Manage your secrets infrastructure via AI by listing, creating, updating, and auditing credentials across environments with end-to-end encryption. It gives you a direct line to your sensitive configuration data, letting you handle everything from a single chat interface. You can query project metadata, browse dev and production environments, and manage service identities without the constant context switching of a manual dashboard. ## Overview - **Category:** developer-tools - **Price:** Free - **Endpoint:** https://edge.vinkius.com/vk_preview_ldGO0JB95xTZUf7eLk52XBp8d2bgs0ppYPfZvUHl/ai-agent-connect - **Tags:** secrets-management, encryption, devops, compliance, env-variables ## Description Infisical lets you manage your secrets infrastructure through natural conversation with your AI client. Instead of jumping between different dashboards to find, update, or audit credentials, you can handle everything from a single chat interface. This Connector acts as a bridge to your sensitive data, allowing you to query project configurations, browse environments like dev or staging, and manage service identities without the usual context switching. When you connect this to your workflow, the manual work of copying and pasting keys disappears. You can ask your agent to pull specific secrets, create new configuration values, or check audit logs for security compliance. It handles the heavy lifting of secret management with end-to-end encryption, ensuring your production environment stays secure while you stay productive. By using this through the Vinkius catalog, you get a way to interact with your most sensitive configuration data using plain English. It turns a series of clicks and browser tabs into a direct conversation, making it easier to manage complex environments without the overhead of a manual UI. Think about the times you've had to hunt for a specific API key in the middle of a deployment. With this connection, those moments of searching and manual entry are gone. Your agent can verify what secrets are available in a given path or give you a high-level overview of your project's metadata. It’s a more fluid way to handle DevOps tasks, moving you away from the dashboard and back into the flow of your actual work. ## Tools ### list_audit_logs List recent audit log entries for the project. This is essential for compliance and security monitoring. ### list_environments List all environments in the configured Infisical project. Use this to discover available environments before switching context. ### create_secret Provision new configuration values or credentials into your project. This helps you set up new keys without manual entry. ### delete_secret Remove a secret from the project permanently. Use this with caution as the action cannot be undone. ### get_project_info Get information about the configured Infisical project. This is useful for checking project metadata. ### get_secret Get a specific secret by its key name. You can optionally specify a path if the secret is not at the root level. ### list_identities List all machine identities configured in the organization. This is useful for auditing automated access. ### list_secrets List all secrets in a given path of the current project and environment. Use this to audit what secrets are available or to verify configuration. ### update_secret Update the value of an existing secret. The key must already exist in the project. ## Prompt Examples **Prompt:** ``` Show me all the secrets in our production environment. ``` **Response:** ``` I've pulled the secrets for the production environment. Here is the list: | Key | Path | Last Updated | | :--- | :--- | :--- | | **DATABASE_URL** | /config/db | 2 hours ago | | **STRIPE_KEY** | /config/payments | 1 day ago | | **REDIS_URL** | /config/cache | 5 days ago | Would you like me to get the specific value for any of these? ``` **Prompt:** ``` Who accessed the secrets in the last hour? ``` **Response:** ``` Here are the recent audit logs for the project from the last hour: * **User:** `deploy-agent-01` | **Action:** `get_secret` | **Key:** `DATABASE_URL` | **Time:** 14:02:10 * **User:** `admin_user` | **Action:** `list_secrets` | **Path:** `/` | **Time:** 14:15:44 Everything looks normal. Do you need a more detailed log for a specific user? ``` **Prompt:** ``` What environments do we have for the web project? ``` **Response:** ``` The web project has the following environments configured: * **Development** (dev) * **Staging** (staging) * **Production** (prod) Which one would you like to switch to? ``` ## Capabilities ### Create new secrets Provision new configuration values or credentials instantly. ### Update existing secrets Change values for your current environment without leaving your chat. ### Audit access logs Check who accessed what and when for security compliance. ### Browse environments Switch context between dev, staging, and production easily. ### Manage machine identities List and organize service identities across your organization. ### List environment secrets Audit what secrets exist in a specific path or project. ## Use Cases ### Updating production credentials during an incident A DevOps engineer needs to update a database password for production but can't find the right dashboard. They ask the agent to use update_secret to change it. ### Performing a security audit for compliance A security auditor needs to check who accessed the Stripe keys last night. They ask the agent to list_audit_logs for the last 24 hours. ### Verifying staging environment configuration A developer is setting up a new staging environment and needs to see what secrets are already there. They use list_secrets to check the root path. ### Auditing machine identities for automated access A platform engineer needs to see all machine identities to audit service accounts. They use list_identities to get a full overview. ## Benefits - Stop switching tabs by using list_secrets to audit configuration directly in your chat. - Reduce human error during deployments by using create_secret to provision keys with precise naming. - Stay compliant by using list_audit_logs to pull security logs for your records instantly. - Simplify environment switching with list_environments so your agent always knows the context. - Manage machine access more efficiently using list_identities to see all service permissions. - Update production values faster with update_secret to modify keys without manual dashboard edits. ## How It Works The bottom line is you get a natural language interface for your secrets infrastructure. 1. Connect your Infisical Service Token and API URL to your AI client. 2. Specify your Project ID and the environment slug you want to target. 3. Ask your agent to perform actions like "list all production secrets" or "update the database key." ## Frequently Asked Questions **Can I use the Infisical MCP to manage my production keys?** Yes, you can use it to manage production keys by connecting it to your Infisical project. It allows your AI agent to list, create, and update keys safely using your existing infrastructure. **How does the Infisical MCP keep my data safe?** It uses your Infisical Service Token and maintains end-to-end encryption. Your secrets stay secure within your Infisical account while you interact with them through your agent. **Can my AI agent update secrets automatically?** Yes, your agent can use the update_secret tool to change secret values. This is great for automating configuration updates without manual dashboard entry. **Does the Infisical MCP support audit logs?** Yes, it can pull recent audit logs for your project. This makes it much easier to monitor access patterns and maintain security compliance. **Can I switch between dev and prod environments?** Yes, you can use the list_environments tool to see what's available and then ask your agent to focus on a specific environment like dev or prod. **Is the Infisical MCP good for DevOps teams?** It is specifically designed for DevOps and security teams. It removes the friction of switching between dashboards and allows you to manage secrets through a natural conversation. **Is it safe to manage secrets through an AI agent?** Your credentials are stored encrypted in the Vinkius vault, transmitted only to the Infisical API at runtime. The AI agent never sees raw secret values in its context — only structured operational results.