"memory" matching MCP servers:

Lets an AI assistant inspect a Linux machine through ten fixed read-only tools covering load, memory, disks, processes, ports, interfaces, systemd units, the journal, and allowlisted log files. All output is size-capped, control-character-stripped and secret-redacted, and nothing in the package can write to the system.

Enables AI coding agents to dynamically inspect running software — API hooks, syscall summaries, memory inspection, and HTTP/TLS/DNS reconstruction from live traffic or pcaps — across Linux, macOS, and Android targets. Every finding is returned as a structured Evidence Record with observation, artifacts, confidence, and explicit limitations, so agents cite proof rather than guess.

Enables AI clients (Codex, Claude, OpenCode, ChatGPT-compatible) to securely administer a YunoHost server, run bounded read-only incident diagnostics, and develop _ynh packages, through Nostr-signed, policy-controlled, fully audited access. Also supports optional delegated short-lived agent identities and an authenticated local memory façade.

Enables MCP clients to run Kali Linux security tooling locally through Docker, covering network scanning, symbol and binary analysis, string extraction, and packet capture analysis. It also provides an isolated sandbox for executing commands with configurable memory, CPU, network, and timeout limits plus file transfer between host and container.