AI agents call get_requirements to retrieve information from Rfcxml without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool performs information retrieval on RFC documents—it parses and returns normative requirement statements (MUST/SHOULD/MAY clauses) without modifying, executing, or destroying any data. The operation is purely read-only with no side effects. It belongs in the Read category with low severity since misuse would only result in incorrect information retrieval, not system compromise or data loss.
From the tool's definition Tool extracts (retrieves) normative requirements from RFC documents in structured format. The description uses the verb 'Extract' which implies reading/querying existing data from RFC documents.
Attacks that exploit this kind of access
Extract normative requirements (MUST/SHOULD/MAY) from RFC in structured format. It is categorised as a Read tool in the Rfcxml MCP Server, which means it retrieves data without modifying state.
Register the Rfcxml MCP server in PolicyLayer and add a rule for get_requirements: 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 Rfcxml. Nothing to install.
get_requirements 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 get_requirements 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 get_requirements. 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.
get_requirements is provided by the Rfcxml MCP server (shuji-bonji/rfcxml-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 →