Decode a JWT header and payload without verifying the signature.
AI agents call decode_jwt to retrieve information from MCP Server Fly without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
JWT decoding is purely informational retrieval; it extracts claims and header information from an encoded token but does not verify, modify, create, delete, or execute anything. The tool performs data extraction similar to other read operations on this server (fetch_url, extract_urls).
From the tool's definition Tool description states it 'Decode[s] a JWT header and payload without verifying the signature' — a read-only operation that retrieves and parses token data without modification, execution, or side effects.
Attacks that exploit this kind of access
Decode a JWT header and payload without verifying the signature. It is categorised as a Read tool in the MCP Server Fly MCP Server, which means it retrieves data without modifying state.
Register the MCP Server Fly MCP server in PolicyLayer and add a rule for decode_jwt: 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 MCP Server Fly. Nothing to install.
decode_jwt 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 decode_jwt 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 decode_jwt. 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.
decode_jwt is provided by the MCP Server Fly MCP server (kushalkachari993-dev/mcp-server-fly). 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 →