The Agent Skills Directory

[SAFE]: No malicious patterns or security risks were detected. The skill performs legitimate binary analysis tasks.
[INDIRECT_PROMPT_INJECTION]: The skill processes strings and pseudocode from untrusted binary files. This is an inherent attack surface for any analysis tool, but since this is the primary purpose of the skill and the capabilities are limited to the analysis platform, it does not represent a significant risk.
Ingestion points: Extracted strings and function pseudocode are read from the target binary in SKILL.md.
Boundary markers: No specific boundary markers or instruction-override warnings are used when processing extracted text.
Capability inventory: Capabilities are limited to rename_function, add_note, and open_function within the Eleven analysis environment.
Sanitization: No sanitization or filtering of binary strings is performed.

first-look