The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to execute various command-line tools including ysoserial.exe, Blacklist3r.exe, and curl to generate payloads and interact with target endpoints.
[EXTERNAL_DOWNLOADS]: It includes steps to install the badsecrets library from PyPI, which is used for identifying weak cryptographic keys.
[DATA_EXFILTRATION]: The skill defines a process for capturing and logging sensitive information from targets, such as credentials and exploitation evidence, to the engagement/evidence/ directory.
[PROMPT_INJECTION]: The skill presents an indirect prompt injection surface through the ingestion of external data.
Ingestion points: External target data such as __VIEWSTATE values, JSON responses, and engagement state retrieved via get_state_summary().
Boundary markers: The instructions do not specify any delimiters or warnings to treat ingested target data as untrusted content.
Capability inventory: The skill has access to command execution (shell tools) and file system writes (logging evidence).
Sanitization: No validation or sanitization routines are defined for the data processed from the target or the state server.

deserialization-dotnet