Run a sequence of commands in a persistent shell session (preserves cwd, env vars).
AI agents invoke omniwire_shell to trigger actions in OmniWire. What it does depends on the arguments the agent supplies, and its effects often reach beyond the immediate call — builds kicked off, notifications sent, workflows started.
This tool executes shell commands on remote infrastructure nodes. Shell execution is inherently Execute-category (not merely Write), as the blast radius depends entirely on what commands an AI agent provides — it could install malware, exfiltrate data, modify system configurations, or trigger destructive operations across multiple nodes.
From the tool's definition Run a sequence of commands in a persistent shell session — directly executes arbitrary shell commands across distributed infrastructure with persistent state (cwd, env vars) via SSH2 connection pooling to multiple nodes.
Documented attack patterns abuse exactly the kind of access omniwire_shell gives an agent:
PolicyLayer is an MCP gateway — it sits between your AI agents and OmniWire, and nothing reaches the server without passing your rules. This is the rule we recommend for omniwire_shell:
{
"version": "1",
"default": "deny",
"tools": {
"omniwire_shell": {
"limits": [
{
"counter": "omniwire_shell_rate",
"window": "minute",
"max": 10,
"scope": "grant"
}
]
}
}
}omniwire_shell stays usable, but rate-capped — a runaway agent can't fire it dozens of times a minute. Everything else on the server is denied unless you say otherwise.
Free to start. No card required.
Run a sequence of commands in a persistent shell session (preserves cwd, env vars). It is categorised as a Execute tool in the OmniWire MCP Server, which means it can trigger actions or run processes. Use rate limits and argument validation.
Register the OmniWire MCP server in PolicyLayer and add a rule for omniwire_shell: allow, deny, rate-limit, or require approval. Point your MCP client at the PolicyLayer proxy URL and the rule is enforced on every call, before it reaches OmniWire. Nothing to install.
omniwire_shell is a Execute tool with high risk. Execute tools should be rate-limited and have argument validation enabled.
Yes. Add a rate_limit block to the omniwire_shell rule in your PolicyLayer policy. For example, setting max: 10 and window: 60 limits the tool to 10 calls per minute. Rate limits are tracked per agent session and reset automatically.
Set action: deny in the PolicyLayer policy for omniwire_shell. The AI agent will receive a policy violation error and cannot call the tool. You can also include a reason field to explain why the tool is blocked.
omniwire_shell is provided by the OmniWire MCP server (voidchecksum/omniwire). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Start from OmniWire, add the rest of your stack, and see everything your agents can call. Then put policy on all of it.
Free to start. No card required.
88 OmniWire tools catalogued and risk-classified — across an index of 43,000+ MCP servers.