AI agents call search_manga to retrieve information from Mangadex without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool retrieves and queries data from MangaDex without creating, modifying, deleting, or executing any operations. It is a straightforward read-only search function that returns search results and metadata. The low severity reflects minimal risk from misuse—an AI agent performing searches poses no danger to data integrity, system operations, or user accounts.
From the tool's definition Tool description states it 'Search[es] MangaDex by title and return[s] MangaDex search metadata only.' The verb 'search' and the clarification that it returns 'metadata only' indicate a query operation with no side effects.
Attacks that exploit this kind of access
Search MangaDex by title and return MangaDex search metadata only. It is categorised as a Read tool in the Mangadex MCP Server, which means it retrieves data without modifying state.
Register the Mangadex MCP server in PolicyLayer and add a rule for search_manga: 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 Mangadex. Nothing to install.
search_manga 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 search_manga 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 search_manga. 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.
search_manga is provided by the Mangadex MCP server (kenhendricks00/mangadex-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 →