what_breaks
AI agents call what_breaks to retrieve information from Git Project Xray without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
Based on the server description mentioning 'assess the impact of changes using ast-grep' and the sibling tools (explore_repo, find_symbol) being read-oriented analysis tools, 'what_breaks' most likely performs static analysis to identify potentially affected code paths. This is a read/query operation. However, the empty description lowers confidence significantly — it could theoretically execute code or make changes.
From the tool's definition Tool name 'what_breaks' suggests impact analysis (finding what breaks given a change); description is empty and uninformative.
Attacks that exploit this kind of access
what_breaks. It is categorised as a Read tool in the Git Project Xray MCP Server, which means it retrieves data without modifying state.
Register the Git Project Xray MCP server in PolicyLayer and add a rule for what_breaks: 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 Project Xray. Nothing to install.
what_breaks 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 what_breaks 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 what_breaks. 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.
what_breaks is provided by the Git Project Xray MCP server (jamie-bitflight/git-project-xray-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 →