This record as markdown: /tools/io-github-ariffazil-aaa-mcp/capability-search.md
What capability_search does on Arifos
AI agents call capability_search to retrieve information from Arifos without modifying anything. It is typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
Why capability_search is rated Low
This tool retrieves and searches information about available MCP tools without creating, modifying, deleting, executing code, or moving funds. It is a passive read operation that discovers metadata about tools in a federated system. The low severity reflects minimal risk: even if misused, a search operation cannot directly harm data or systems.
From the tool's definition Tool description states it performs 'Semantic search over the federation's 97 MCP tools' — a query/retrieval operation with no modification or execution of external systems.
Attacks that exploit this kind of access
The rule that runs capability_search safely
PolicyLayer is an MCP gateway: it sits between your AI agents and Arifos, and checks every tool call against a rule you set before the call runs. Nothing changes on the server itself. For capability_search, this is the rule to start with:
capability_search is read-only, so it stays allowed. Everything else on the server is denied unless you say otherwise.
The button opens the PolicyLayer dashboard: create your workspace, connect Arifos, apply this rule, and every capability_search call is checked against it from then on.
Questions about capability_search
Semantic search over the federation's 97 MCP tools. It is categorised as a Read tool in the Arifos MCP Server, which means it retrieves data without modifying state.
Register the Arifos MCP server in PolicyLayer and add a rule for capability_search: 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 Arifos. Nothing to install.
capability_search 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 capability_search 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 capability_search. 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.
capability_search is provided by the Arifos MCP server (arifos). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
More on Arifos, and thousands of servers like it.
This server
Across the catalogue