The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill fetches application metadata from Apple's official iTunes Search API and retrieves configuration manifests from specified target domains during the discovery phase.
[COMMAND_EXECUTION]: Analysis of the automated alert regarding remote code execution via python3 confirms it is a false positive. The command python3 -m json.tool is used to format JSON data received from the Apple iTunes API, which is a built-in and safe data processing method.
[INDIRECT_PROMPT_INJECTION]: The skill has an attack surface for indirect prompt injection as it processes data extracted from third-party mobile applications.
Ingestion points: The skill analyzes output from strings and plutil commands derived from application binaries and configuration files (SKILL.md).
Boundary markers: None; the skill operates directly on the extracted data strings.
Capability inventory: Includes grep, find, and curl across various stages to search for secrets and configuration weaknesses (SKILL.md).
Sanitization: None; the data is treated as raw text for pattern matching.

ios-redteam-pipeline