skills/mukul975/anthropic-cybersecurity-skills/analyzing-network-covert-channels-in-malware/Gen Agent Trust Hub
analyzing-network-covert-channels-in-malware
Pass
Audited by Gen Agent Trust Hub on Sep 14, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to process untrusted external network traffic data (PCAP files) which represents a potential attack surface.
- Ingestion points: Analysis scripts in
SKILL.mdandscripts/agent.pyingest packet capture files provided by the user via thescapy.all.rdpcapfunction. - Boundary markers: The skill does not employ explicit boundary markers as the parsed data is analyzed for statistical anomalies rather than being interpolated into complex prompts.
- Capability inventory: The skill's capabilities are limited to statistical analysis (entropy and volume checks) and outputting results to the terminal. It lacks the ability to execute commands, write to the filesystem, or initiate network connections based on the content of the analyzed packets.
- Sanitization: The scripts use safe decoding practices, such as
errors='replace', when handling potentially malformed strings from network protocols like DNS.
Audit Metadata