The Agent Skills Directory

[DYNAMIC_CONTEXT_INJECTION]: The skill utilizes dynamic context injection (!command) to check the status of the local environment when the skill is loaded. These commands check for the presence of the aircrack-ng binary and verify if specific wrapper scripts are available locally.
Evidence: !command -v aircrack-ng > /dev/null 2>&1 && echo "YES -- $(aircrack-ng --version 2>&1 | head -1)" || echo "NO -- Install..." (SKILL.md)
Evidence: !test -f scripts/aircrack-ng/analyze-wireless-networks.sh && echo "YES -- use wrapper scripts for structured JSON output" || echo "NO -- using standalone mode with direct commands" (SKILL.md)
[INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from the wireless environment, including SSIDs and other network metadata, which could contain instructions meant to influence the agent's behavior.
Ingestion points: airodump-ng scanning and analyze-wireless-networks.sh (SKILL.md)
Boundary markers: None identified in the provided instructions.
Capability inventory: Shell execution of the aircrack-ng suite, file system access for saving captures, and execution of local bash scripts.
Sanitization: No explicit sanitization or filtering of captured WiFi metadata is described.
[COMMAND_EXECUTION]: The skill instructs the agent to execute shell commands using the aircrack-ng suite and local bash scripts for wireless analysis, monitor mode management, and dictionary attacks. Commands like airmon-ng check kill interact with system processes.

aircrack-ng