The Agent Skills Directory

[COMMAND_EXECUTION]: The skill provides multiple shell command templates for executing the adinfo utility to collect information from domain environments.- [CREDENTIALS_UNSAFE]: The usage guidelines instruct the user or agent to pass domain passwords and NTLM hashes directly as command-line arguments (e.g., -p password and -H NTLM_HASH), which exposes sensitive credentials to local logging and process monitoring tools.- [EXTERNAL_DOWNLOADS]: The skill promotes the use of a tool hosted on a third-party GitHub repository (github.com/lzzbb/Adinfo) that does not belong to a trusted organization or well-known service.- [PROMPT_INJECTION]: The skill creates a surface for indirect prompt injection through its information-gathering workflow:
Ingestion points: Untrusted data enters the agent context when it reads the output of the adinfo tool, which parses objects from an Active Directory environment.
Boundary markers: The instructions do not define delimiters to separate the tool's output from the agent's internal logic or following instructions.
Capability inventory: The agent is authorized to execute follow-on shell commands (e.g., GetUserSPNs.py, GetNPUsers.py) based on the gathered information.
Sanitization: There is no implementation of sanitization or validation for the strings retrieved from the external AD environment before they are processed by the agent.

adinfo-enum