Get episodes from the graph memory.
AI agents call get_episodes to retrieve information from Graphiti MCP Server without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool retrieves episodes from the knowledge graph without creating, modifying, or deleting data. It has no side effects and returns information only. While the server itself supports write operations (add_memory, delete_entity_edge, delete_episode, clear_graph), this particular tool is a straightforward read operation with minimal blast radius if misused by an AI agent.
From the tool's definition Tool name is 'get_episodes' and description states 'Get episodes from the graph memory' — both indicate data retrieval with no modification. The 'get_' prefix and verb 'Get' confirm a read-only operation.
Attacks that exploit this kind of access
Get episodes from the graph memory. It is categorised as a Read tool in the Graphiti MCP Server MCP Server, which means it retrieves data without modifying state.
Register the Graphiti MCP Server MCP server in PolicyLayer and add a rule for get_episodes: 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 Graphiti MCP Server. Nothing to install.
get_episodes 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 get_episodes 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 get_episodes. 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.
get_episodes is provided by the Graphiti MCP Server MCP server (tesfandiari1/graphiti-mcp-server). 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 →