The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill provides instructions for installing Tailscale using a remote script execution pattern.
Evidence: curl -fsSL https://tailscale.com/install.sh | sh in references/vpn.md.
Context: Tailscale is a well-known service, and this command is the standard installation method for its Linux client. No malicious parameters or unknown sources were detected.
[COMMAND_EXECUTION]: The skill requires and provides instructions for executing system-level commands to manage networking, firewalls, and services.
Evidence: Commands such as ip link, nft list ruleset, systemctl enable nftables, and sysctl -w net.ipv4.ip_forward=1 are found throughout the documentation.
Context: These capabilities are essential for the primary purpose of configuring Linux networking and troubleshooting.
[PROMPT_INJECTION]: The skill exhibits an attack surface for Indirect Prompt Injection (Category 8) due to its use of network diagnostic tools that ingest untrusted data.
Ingestion points: Network packets and headers captured via tcpdump and tshark, as well as HTTP responses fetched via curl (referenced in references/troubleshooting.md).
Boundary markers: Absent; the skill does not specify delimiters or instructions to ignore embedded content in tool output.
Capability inventory: Full access to networking configuration tools (ip, nft, wg, cloudflared) and service management (systemctl) in SKILL.md and reference files.
Sanitization: Absent; there is no explicit validation or escaping of the output from diagnostic tools before it is processed by the agent.

networking