The Agent Skills Directory

[DYNAMIC_CONTEXT_INJECTION]: The skill executes a shell command at load time to perform environment detection, checking for the existence of local wrapper scripts.
Evidence: !test -f scripts/nmap/discover-live-hosts.sh && echo "YES" || echo "NO"`` in SKILL.md.
[COMMAND_EXECUTION]: The skill performs shell execution of networking and system tools to carry out its intended reconnaissance functions.
Evidence: Direct usage of nmap, dig, curl, gobuster, and local bash scripts in SKILL.md (e.g., bash scripts/nmap/discover-live-hosts.sh $ARGUMENTS -j -x).
[INDIRECT_PROMPT_INJECTION]: The skill ingests data from untrusted external network services (such as DNS records and service version strings) which could theoretically contain malicious instructions targeting the agent.
Ingestion points: Captured output from nmap, dig, curl, and gobuster commands (SKILL.md).
Boundary markers: Absent; the agent is instructed to review tool output directly without specific delimiters or instructions to ignore embedded content.
Capability inventory: Shell command execution and network tool invocation across multiple steps in SKILL.md.
Sanitization: Absent; the skill does not specify any sanitization or filtering of the output received from external network scans before the agent processes it.

recon