The Agent Skills Directory

[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because its core functionality involves the agent ingesting and analyzing untrusted data from external binaries.
Ingestion points: The skill instructs the agent to read and process potentially malicious files (e.g., sample.exe) using tools such as pefile, strings, floss, and capa in SKILL.md.
Boundary markers: There are no explicit instructions or delimiters provided to help the agent distinguish between its own system instructions and the potentially adversarial text extracted from analyzed malware samples.
Capability inventory: The skill utilizes shell command execution and Python scripting to perform deep file analysis and environment manipulation.
Sanitization: No mechanisms for sanitizing or escaping the output of the analysis tools (which often contain raw strings from malware) are specified before the data is returned to the agent's context.

malware-analysis