Search for commands in shell history
AI agents call search_commands to retrieve information from Mcp Histfile without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
The tool searches and retrieves historical data from shell history without executing commands, modifying history, or causing side effects. This is a pure Read operation. Severity is low because accessing command history poses minimal risk compared to executing or destructive operations, though exposure of sensitive commands in history is a minor information disclosure concern.
From the tool's definition Tool description states 'Search for commands in shell history' - a query/search operation with no modification or execution capability. Sibling tools include 'get_command' and 'get_recent_commands', all consistent with retrieval-only functionality.
Attacks that exploit this kind of access
Search for commands in shell history. It is categorised as a Read tool in the Mcp Histfile MCP Server, which means it retrieves data without modifying state.
Register the Mcp Histfile MCP server in PolicyLayer and add a rule for search_commands: 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 Mcp Histfile. Nothing to install.
search_commands 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 search_commands 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 search_commands. 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.
search_commands is provided by the Mcp Histfile MCP server (rajpdus/mcp-histfile). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Every MCP server has a record like this.
Type a name, get the same breakdown: verified identity, auth posture, risk grade, capabilities, recommended policy.
Teams ship this data inside their own products. See what a licence covers →