Measure semantic alignment between prompt and outputs using embeddings
AI agents call prompt.assess_relevance to retrieve information from ContextForge MCP Gateway without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool retrieves and processes data (embeddings and similarity metrics) without modifying, deleting, or executing code. It is purely an analytical read operation that assesses semantic similarity between inputs and outputs. No data is created, destroyed, or irreversibly changed.
From the tool's definition Tool 'prompt.assess_relevance' measures semantic alignment using embeddings—a comparison/analysis operation with no state changes, deletions, or external side effects.
Attacks that exploit this kind of access
Measure semantic alignment between prompt and outputs using embeddings. It is categorised as a Read tool in the ContextForge MCP Gateway MCP Server, which means it retrieves data without modifying state.
Register the ContextForge MCP Gateway MCP server in PolicyLayer and add a rule for prompt.assess_relevance: 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 ContextForge MCP Gateway. Nothing to install.
prompt.assess_relevance 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 prompt.assess_relevance 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 prompt.assess_relevance. 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.
prompt.assess_relevance is provided by the ContextForge MCP Gateway MCP server (jrmatherly/mcp-context-forge). 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 →