$ loading_
Securely lets AI use real secrets with human approval per release.
Copy the install command and let the AI configure it · recommended for beginners
No copy-paste install info for "stillvault-mcp" yet — see the docs or source repo.
Use stillvault-mcp to connect to the production database and run a read-only query for yesterday's total orders. Before requesting the database password release, show me the purpose, SQL query, and risk notes, then wait for my approval before continuing.
It first presents the secret release request and query details, then after human approval runs the query securely and returns the results.
Use stillvault-mcp to call the payment service API and check failed transactions from the last 24 hours. Do not expose the API key directly; first create a secret release request for my approval, explaining the scope and expected response fields.
It creates an approvable secret-use request and, once approved, performs the API call and returns a summary of failed transactions.
Before deploying to production, use stillvault-mcp to obtain cloud platform credentials. Every credential release must be confirmed by me and include the reason, target environment, and deployment steps.
It outputs a deployment plan with an approval-gated credential request, then deploys only after approval, keeping credentials out of the model context.
Securely store and manage encrypted credentials for AI agents without hardcoded keys.
Securely store and retrieve local secrets for AI workflows without sending them to LLM APIs.
Securely manage secrets for AI coding agents without exposing them to LLM context.
Securely manage secrets from keychains or encrypted files without chat exposure.
Securely let AI agents use stored credentials without exposing plaintext secrets.
Securely inject secrets into agent commands without exposing them to models.