Generate embeddings via the gateway (default backend: local Ollama nomic-embed-text).
AI agents call gateway.embed to retrieve information from Adaptive LLM Gateway without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
Generating embeddings is a read/query operation that transforms input text into vector representations. It retrieves computed data (embedding vectors) without modifying any stored state, executing commands, or incurring direct financial transactions. The local Ollama backend reduces blast radius further.
From the tool's definition Generate embeddings via the gateway (default backend: local Ollama nomic-embed-text)
Attacks that exploit this kind of access
Generate embeddings via the gateway (default backend: local Ollama nomic-embed-text). It is categorised as a Read tool in the Adaptive LLM Gateway MCP Server, which means it retrieves data without modifying state.
Register the Adaptive LLM Gateway MCP server in PolicyLayer and add a rule for gateway.embed: 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 Adaptive LLM Gateway. Nothing to install.
gateway.embed 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 gateway.embed 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 gateway.embed. 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.
gateway.embed is provided by the Adaptive LLM Gateway MCP server (renefichtmueller/adaptive-llm-gateway). 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 →