Decode base64 to UTF-8 string.
AI agents call base64_decode to retrieve information from Nexus Core without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
base64_decode performs reversible string decoding only. It reads encoded input and returns decoded output with no state changes, data creation, modification, or deletion. This is a simple utility function with minimal blast radius if misused by an agent—worst case, an attacker gains insight into previously base64-encoded content, but cannot alter systems or data.
From the tool's definition Tool description states 'Decode base64 to UTF-8 string'—a pure data transformation that retrieves/decodes information without modification, deletion, or side effects.
Attacks that exploit this kind of access
Decode base64 to UTF-8 string. It is categorised as a Read tool in the Nexus Core MCP Server, which means it retrieves data without modifying state.
Register the Nexus Core MCP server in PolicyLayer and add a rule for base64_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 Nexus Core. Nothing to install.
base64_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 base64_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 base64_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.
base64_decode is provided by the Nexus Core MCP server (noumenon-ai/nexus-core). 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 →