The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill downloads and executes the official Tailscale installation script from tailscale.com. This is a well-known service providing networking and security software.
[COMMAND_EXECUTION]: The helper scripts scripts/setup_exit_node.sh and scripts/setup_subnet_router.sh use the eval command to execute shell strings built from positional arguments. This approach is vulnerable to command injection if untrusted or malicious input is passed as arguments to these scripts.
[PRIVILEGE_ESCALATION]: The skill's setup scripts perform administrative system configurations, including enabling IP forwarding through sysctl and modifying configuration files in /etc/sysctl.d/. These operations require and request root privileges to function.
[INDIRECT_PROMPT_INJECTION]: 1. Ingestion points: The skill processes external inputs such as subnet CIDR ranges, authentication keys, and ACL policy definitions in JSON format. 2. Boundary markers: There are no explicit delimiters or instructions provided to the agent to treat these inputs as data rather than instructions. 3. Capability inventory: The skill has the ability to execute shell commands, perform system-wide network configuration, and interface with the Tailscale API. 4. Sanitization: The scripts do not implement validation or escaping for the arguments before they are executed via eval or written to system configuration paths.

tailscale