The Agent Skills Directory

[COMMAND_EXECUTION]: The skill provides numerous commands for network forensics and protocol analysis using standard utilities such as tshark, tcpdump, nmap, masscan, and strings to analyze PCAP files.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process untrusted external data in the form of network capture (PCAP) files. As it parses these files using tshark, strings, and scapy without explicit sanitization or boundary markers, it is theoretically susceptible to indirect prompt injection if a packet payload contains instructions designed to influence the agent's behavior.
Ingestion points: Reads PCAP files via tshark, rdpcap, and strings (e.g., SKILL.md, reference/protocols.md).
Boundary markers: None identified.
Capability inventory: The skill uses Bash, Write, and specialized networking tools to perform analysis and potentially output results to the filesystem.
Sanitization: None identified; raw packet data is processed directly.
[EXTERNAL_DOWNLOADS]: The instructions recommend the installation of established networking tools and libraries, including scapy via pip and wireshark/nmap via brew. These are well-known, legitimate tools for the stated purpose of network analysis.

networking