Find ChatGPT conversations whose title contains query (substring match).
AI agents call search_conversations to retrieve information from Codex Chats 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 existing conversation data without modifying, deleting, or executing any operations. It is a standard search/filter operation with no side effects, fitting the 'Read' category. Severity is low because the blast radius of misuse is minimal—an agent can only discover conversations that already exist, with no capability to alter state or access sensitive external systems.
From the tool's definition Tool performs a substring search on conversation titles to find matching conversations. Description explicitly states 'Find' with no modification or deletion of data.
Attacks that exploit this kind of access
Find ChatGPT conversations whose title contains query (substring match). It is categorised as a Read tool in the Codex Chats MCP Server, which means it retrieves data without modifying state.
Register the Codex Chats MCP server in PolicyLayer and add a rule for search_conversations: 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 Codex Chats. Nothing to install.
search_conversations 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_conversations 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_conversations. 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_conversations is provided by the Codex Chats MCP server (shoyu-ramen/codex-chats-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 →