The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes several command-line tools such as nc, nmap, and msfconsole to perform service discovery and data extraction from remote Memcached instances.
[PRIVILEGE_ESCALATION]: Instructions include the use of sudo apt install to set up the environment by installing libmemcached-tools and php-memcached on the host system.
[DYNAMIC_EXECUTION]: The skill generates and executes inline PHP code (php -r) and shell pipelines to interact with the Memcached protocol and process its output.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and display untrusted data retrieved from external Memcached servers (such as session tokens or serialized objects), which could contain malicious instructions for the agent.
Ingestion points: Data is retrieved from remote servers via nc, memccat, and PHP scripts in both SKILL.md and references/memcache-techniques.md.
Boundary markers: The instructions lack explicit boundary markers or warnings to the agent to treat the retrieved cache content as untrusted data.
Capability inventory: The agent environment includes capabilities for shell command execution, network communication, and system-level package installation.
Sanitization: There are no provided mechanisms or instructions for sanitizing the retrieved data before it is processed by the agent's logic.

memcache-pentesting