AI agents call prime to retrieve information from Sandy without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
The tool simply reads and returns the contents of a SKILL.md file, which is a documentation/instructions file. This is a pure read operation with no writes, execution, or destructive actions. Severity is low as the worst case is exposure of internal documentation.
From the tool's definition 'Return the MCP SKILL.md content' — retrieves static documentation/markdown file content with no side effects
Attacks that exploit this kind of access
Return the MCP SKILL.md content. It is categorised as a Read tool in the Sandy MCP Server, which means it retrieves data without modifying state.
Register the Sandy MCP server in PolicyLayer and add a rule for prime: 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 Sandy. Nothing to install.
prime 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 prime 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 prime. 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.
prime is provided by the Sandy MCP server (jamestelfer/sandy). 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 →