Decode a JWT without verifying its signature. POST { token }. Returns the decoded header and payload, plus issuedAt/expiresAt/notBefore as ISO timestamps, and expired / notYetValid flags. Signature is NOT checked — decode/inspection only. For agents reading token claims (scopes, sub, exp) before ...
AI agents call dev.jwt-decode to retrieve information from Mcp without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool retrieves and inspects JWT token metadata without side effects. It performs cryptographic inspection/validation (decoding) but not enforcement (signature verification is skipped). The output is purely informational—token claims, timestamps, validity flags—suitable for agents to make decisions before acting. No modification, deletion, or external execution occurs.
From the tool's definition Tool decodes JWT tokens and returns header, payload, and timestamp fields. Explicitly states 'Signature is NOT checked — decode/inspection only' and is intended 'For agents reading token claims'. No data is modified, created, deleted, or executed.
Attacks that exploit this kind of access
Decode a JWT without verifying its signature. POST { token }. Returns the decoded header and payload, plus issuedAt/expiresAt/notBefore as ISO timestamps, and expired / notYetValid flags. Signature is NOT checked — decode/inspection only. For agents reading token claims (scopes, sub, exp) before acting. It is categorised as a Read tool in the Mcp MCP Server, which means it retrieves data without modifying state.
Register the MCP server in PolicyLayer and add a rule for dev.jwt-decode: 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. Nothing to install.
dev.jwt-decode 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 dev.jwt-decode 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 dev.jwt-decode. 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.
dev.jwt-decode is provided by the MCP server (@2sio/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 →