The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes command-line tools to interact with the environment and the Sealos platform.
Executes scripts/analyze-project-database.mjs to scan the local project for database configurations.
Executes sealos-cli for provisioning and managing cloud databases.
[EXTERNAL_DOWNLOADS]: The skill dynamically acquires necessary tooling from external repositories.
Uses npx -y sealos-cli@latest to download and run the Sealos CLI if it is not present on the host system.
[DATA_EXFILTRATION]: The skill interacts with sensitive files containing credentials.
Reads and modifies project environment files like .env and .env.local to inject database connection strings.
Accesses the user's Sealos configuration at ~/.sealos/auth.json and ~/.sealos/kubeconfig for authentication.
[PROMPT_INJECTION]: The project analyzer script parses contents from the working directory, creating a surface for indirect instructions.
Ingestion: scripts/analyze-project-database.mjs reads project files to detect database signals.
Boundaries: SKILL.md directs the agent to inspect files before editing.
Capabilities: The skill can modify .env files and perform database operations via sealos-cli.
Sanitization: The analyzer extracts only keys and patterns, avoiding full content exposure.

sealos-database