Validate a shell command for safety before execution. Checks for dangerous patterns including: - Destructive commands (rm -rf, format, etc.) - Privilege escalation (sudo, chmod 777, etc.) - Network exfiltration (curl to suspicious URLs, etc.) - Code injection (eval, exec with variables, etc.) A...
Accepts freeform code/query input (command)
Part of the Mund MCP server. Enforce policies on this tool with Intercept, the open-source MCP proxy.
AI agents call mund_validate_command to retrieve information from Mund without modifying any data. This is common in research, monitoring, and reporting workflows where the agent needs context before taking action. Because read operations don't change state, they are generally safe to allow without restrictions -- but you may still want rate limits to control API costs.
Even though mund_validate_command only reads data, uncontrolled read access can leak sensitive information or rack up API costs. An agent caught in a retry loop could make thousands of calls per minute. A rate limit gives you a safety net without blocking legitimate use.
Read-only tools are safe to allow by default. No rate limit needed unless you want to control costs.
tools:
mund_validate_command:
rules:
- action: allowSee the full Mund policy for all 23 tools.
Agents calling read-class tools like mund_validate_command have been implicated in these attack patterns. Read the full case and prevention policy for each:
Other tools in the Read risk category across the catalogue. The same policy patterns (rate-limit, allow) apply to each.
Validate a shell command for safety before execution. Checks for dangerous patterns including: - Destructive commands (rm -rf, format, etc.) - Privilege escalation (sudo, chmod 777, etc.) - Network exfiltration (curl to suspicious URLs, etc.) - Code injection (eval, exec with variables, etc.) Args: - command (string): Shell command to validate (required) - tool_name (string): Name of tool executing command (optional) - agent_id (string): ID of requesting agent (optional) Returns: JSON object with: - command: The validated command - safe: Boolean indicating if command is safe - risk_level: 'safe' | 'suspicious' | 'dangerous' - warnings: Array of warning messages - blocked: Whether execution was blocked. It is categorised as a Read tool in the Mund MCP Server, which means it retrieves data without modifying state.
Add a rule in your Intercept YAML policy under the tools section for mund_validate_command. You can allow, deny, rate-limit, or validate arguments. Then run Intercept as a proxy in front of the Mund MCP server.
mund_validate_command is a Read tool with low risk. Read-only tools are generally safe to allow by default.
Yes. Add a rate_limit block to the mund_validate_command rule in your Intercept 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 Intercept policy for mund_validate_command. 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.
mund_validate_command is provided by the Mund MCP server (@weave_protocol/mund). Intercept sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Open source. One binary. Zero dependencies.
npx -y @policylayer/intercept