The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes the x shodan utility to perform network scans and host lookups. These commands are localized to the tool's intended reconnaissance functions.
[CREDENTIALS_UNSAFE]: The instructions guide users to set up their own Shodan API keys using an initialization command. This is a secure practice that avoids embedding secrets directly in the skill code.
[PROMPT_INJECTION]: The skill features an AI-powered summary tool (x shodan ::) that processes external data from the Shodan database. This represents an indirect prompt injection surface where the model could potentially be influenced by content found in search results.
Ingestion points: Shodan search results and network intelligence reports from SKILL.md and SKILL.zh.md.
Boundary markers: Absent; there are no explicit delimiters defined to separate untrusted Shodan data from agent instructions.
Capability inventory: Execution of network search and scan commands via the x toolset.
Sanitization: Not defined; the skill does not explicitly detail any filtering or validation of the retrieved Shodan data.

shodan