Search conversations by git repository, branch, or commit
AI agents call search_by_git to retrieve information from Claude Code History MCP 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 conversation data based on git metadata filters. It performs a query operation with no side effects, data modification, code execution, or resource consumption. The blast radius of misuse is limited to information disclosure of historical conversations.
From the tool's definition Tool description states 'Search conversations by git repository, branch, or commit' — a retrieval operation with no modification or execution.
Attacks that exploit this kind of access
Search conversations by git repository, branch, or commit. It is categorised as a Read tool in the Claude Code History MCP MCP Server, which means it retrieves data without modifying state.
Register the Claude Code History MCP server in PolicyLayer and add a rule for search_by_git: 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 Claude Code History MCP. Nothing to install.
search_by_git 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_by_git 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_by_git. 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_by_git is provided by the Claude Code History MCP server (tim0120/claude-code-history-mcp). 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 →