The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill fetches the official installation script from Tailscale's domain (tailscale.com) and provides instructions to download the Headscale binary from its GitHub repository. These are recognized sources for the software being implemented.
[COMMAND_EXECUTION]: The monitoring script in scripts/process.py executes the tailscale CLI to retrieve status information in JSON format for health monitoring.
[PRIVILEGE_ESCALATION]: Instructions include the use of sudo for repository configuration, service initialization, and modifying system networking parameters like IP forwarding, which are standard administrative requirements for VPN deployment.
[INDIRECT_PROMPT_INJECTION]: The audit scripts ingest data from external sources that could potentially contain malicious instructions.
Ingestion points: scripts/agent.py reads data from the Tailscale API, and scripts/process.py reads output from the tailscale CLI.
Boundary markers: The scripts rely on structured JSON parsing without explicit boundary delimiters for natural language content.
Capability inventory: The skill scripts possess capabilities for network requests (requests.get) and local command execution (subprocess.run).
Sanitization: No specific content sanitization or validation is applied to the data retrieved from the API or CLI beyond standard JSON parsing.

deploying-tailscale-for-zero-trust-vpn