AI agents call list_commits_since_last_tag to retrieve information from Git MCP without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool performs pure data retrieval by querying commit history. It does not create, modify, delete, or execute any operations on repositories. The blast radius of misuse is minimal — an agent could only retrieve information already present in the repository, not cause any changes or trigger external effects.
From the tool's definition Tool description states it 'List commit messages since main HEAD and the last git tag' — a read-only operation that retrieves and queries existing git history with no side effects.
Attacks that exploit this kind of access
List commit messages since main HEAD and the last git tag. It is categorised as a Read tool in the Git MCP MCP Server, which means it retrieves data without modifying state.
Register the Git MCP server in PolicyLayer and add a rule for list_commits_since_last_tag: 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 Git MCP. Nothing to install.
list_commits_since_last_tag 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 list_commits_since_last_tag 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 list_commits_since_last_tag. 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.
list_commits_since_last_tag is provided by the Git MCP server (kjozsa/git-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 →