The Agent Skills Directory

[SAFE]: The skill provides legitimate instructions and tools for deploying and managing HashiCorp Vault. All examples use clear placeholders for credentials, and the included Python script follows security best practices for secret handling.
[EXTERNAL_DOWNLOADS]: The skill and reference documentation recommend installing official, well-known libraries (hvac, requests) from public package registries. These references are documented neutrally as standard requirements for the skill's functionality.
[COMMAND_EXECUTION]: The scripts/agent.py utility provides administrative commands to interact with a Vault instance. The script includes a safety feature that masks secret values in its output, reducing the risk of accidental exposure during use.
[INDIRECT_PROMPT_INJECTION]: The skill identifies a surface for data ingestion where content from a Vault server is brought into the agent's context.
Ingestion points: scripts/agent.py retrieves secret data and metadata using the hvac client.
Boundary markers: No specific boundary markers are used in the raw console output of the script.
Capability inventory: The script has capabilities to read, write, list, and delete secrets, as well as perform audit sweeps.
Sanitization: The script implements a masking function that hides all but the first three characters of string values, providing a layer of protection against accidental data leakage.

implementing-secrets-management-with-vault