Decompiles a Java class from a package name
AI agents call decompile-from-package to retrieve information from Mcp Javadc without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
Decompilation is a read-only operation: it takes compiled bytecode and reconstructs source code for inspection. There are no writes, executions, or destructive side effects. The slight risk is that it could expose proprietary code, but the blast radius of misuse is low.
From the tool's definition 'Decompiles a Java class from a package name' — the tool reads/retrieves decompiled source code from an existing Java package, producing output without modifying or deleting anything.
Attacks that exploit this kind of access
Decompiles a Java class from a package name. It is categorised as a Read tool in the Mcp Javadc MCP Server, which means it retrieves data without modifying state.
Register the Mcp Javadc MCP server in PolicyLayer and add a rule for decompile-from-package: 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 Javadc. Nothing to install.
decompile-from-package 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 decompile-from-package 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 decompile-from-package. 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.
decompile-from-package is provided by the Mcp Javadc MCP server (@idachev/mcp-javadc). 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 →