Exec Sandbox
Single exec tool running caller Python in a network-isolated locked-down sandbox.
Open the source (opens github.com) https://github.com/inhuman/mcp-exec
Before you use it
- Runs on your computer with your permissions. Read what it can do before connecting it to an agent.
- Needs credentials: MCP_EXEC_AUTH_TOKEN. Use a key with the least access you can.
What it is
MCP servers from io.github.inhuman.
The problem it solves and who it suits haven't been reviewed by InnerVault. The description above is the source's own.
How to get it
oci package (stdio)
docker.io/idconstruct/mcp-exec:v0.5.4
From the server's entry in the official MCP Registry
Settings it asks for
MCP_EXEC_TRANSPORT· MCP transport. Must be 'stdio' for direct docker/stdio use.MCP_EXEC_DEFAULT_TIMEOUT_S· Default wall-clock timeout per exec, seconds.MCP_EXEC_MAX_TIMEOUT_S· Maximum wall-clock timeout a caller may request, seconds.MCP_EXEC_MAX_OUTPUT_BYTES· Cap on combined stdout/stderr before truncation.MCP_EXEC_MAX_STDIN_BYTES· Cap on stdin passed to the sandboxed code.MCP_EXEC_AUTH_TOKENsecret · Optional X-MCP-AUTH token (http/sse transports only).
Related
-
Python Exec Sandbox
MCP serversSandboxed Python execution for AI agents. PEP 723 inline deps, multi-version Python, zero pollution.
mcp-python-exec-sandbox@0.1.5 -
agent-exec
MCP serversAccountless pay-per-use sandbox, storage, locks, attestation, email inbox and DNS names via x402.
-
agent-exec
MCP serversAccountless pay-per-use sandbox execution, storage, locks, attestation and handoff; pay via x402.
-
Sandbox Env MCP
MCP serversPersistent Docker/SSH sandbox for AI agents: shell exec, file ops, audit log.
sandbox-env-mcp@0.3.1 -
Sprites MCP Server
MCP serversManage Sprites: sandboxed compute environments with exec, services, and checkpoints.
-
ironclaw
MCP serversSandboxed shell exec for MCP clients: run untrusted agent commands in a gVisor container.
ghcr.io/ironsecco/ironclaw-mcp:v0.1.569