AI agents call policy_eval to retrieve information from Git Steer without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This is a read-only security audit tool. It evaluates repositories against existing policies and reports compliance status. No data is created, modified, deleted, or executed—it purely retrieves and assesses information about security policy states (Dependabot, secret scanning, branch protection, advanced security settings).
From the tool's definition Tool name 'policy_eval' and description 'Evaluate managed repos against security policies... Reports pass/fail per control' indicate a query/assessment operation that retrieves security policy compliance status without modifying, executing, or deleting…
Attacks that exploit this kind of access
Evaluate managed repos against security policies (Dependabot, secret scanning, branch protection, advanced security). Reports pass/fail per control. It is categorised as a Read tool in the Git Steer MCP Server, which means it retrieves data without modifying state.
Register the Git Steer MCP server in PolicyLayer and add a rule for policy_eval: 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 Steer. Nothing to install.
policy_eval 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 policy_eval 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 policy_eval. 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.
policy_eval is provided by the Git Steer MCP server (ry-ops/git-steer). 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 →