diff
AI agents call diff to retrieve information from BlobGuard MCP without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
In the context of a blob storage and diff server for code edits, a 'diff' tool most likely computes and returns the difference between two blobs — a read/query operation with no side effects. However, the empty description lowers confidence. If it applied diffs (patched blobs), it would be Write.
From the tool's definition Tool name is 'diff'; description is empty and uninformative.
Attacks that exploit this kind of access
diff. It is categorised as a Read tool in the BlobGuard MCP MCP Server, which means it retrieves data without modifying state.
Register the BlobGuard MCP server in PolicyLayer and add a rule for diff: 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 BlobGuard MCP. Nothing to install.
diff 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 diff 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 diff. 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.
diff is provided by the BlobGuard MCP server (voidious/blobguard-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 →