Fetch messages for a target with id > since_id.
AI agents call fetch_messages to retrieve information from Claude ↔ Codex MCP Relay without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool retrieves messages from a messaging channel without modifying, creating, or deleting any data. It is a straightforward query operation with no side effects. The filtering parameter 'since_id' suggests pagination/historical retrieval rather than any state-changing operation. Severity is low because message retrieval poses minimal risk even if misused by an agent.
From the tool's definition Tool name 'fetch_messages' and description 'Fetch messages for a target with id > since_id' indicate retrieval of existing data with no modification or deletion capability.
Attacks that exploit this kind of access
Fetch messages for a target with id > since_id. It is categorised as a Read tool in the Claude ↔ Codex MCP Relay MCP Server, which means it retrieves data without modifying state.
Register the Claude ↔ Codex MCP Relay MCP server in PolicyLayer and add a rule for fetch_messages: 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 Claude ↔ Codex MCP Relay. Nothing to install.
fetch_messages 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 fetch_messages 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 fetch_messages. 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.
fetch_messages is provided by the Claude ↔ Codex MCP Relay MCP server (wiresharks/claudecodex). 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 →