The Agent Skills Directory

[DYNAMIC_EXECUTION]: The skill uses Python heredocs to execute embedded scripts for processing JSON data and formatting the output. This is a standard practice for including complex logic within shell-based skills and uses fixed templates.
[INDIRECT_PROMPT_INJECTION]: The skill processes external data retrieved from the VirusTotal API. While the data is parsed as structured JSON, the content of fields such as malware tags or vendor descriptions is displayed to the user without explicit character sanitization.
Ingestion points: VirusTotal API response processed in SKILL.md.
Boundary markers: None present for the displayed API content.
Capability inventory: Network requests via curl, file read/write in /tmp, and execution of python3 and system hash utilities.
Sanitization: String values from the API response are parsed but not explicitly escaped before being printed to the console.
[COMMAND_EXECUTION]: Executes standard system utilities including md5sum, sha1sum, and sha256sum for local file hashing, and curl for API communication. These are appropriate for the skill's stated purpose.

malware-hash