Initialize a new memory session and load context. Call once at the start of each conversation.
AI agents use start_session to create or update resources in Claude Memory — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your Claude Memory environment.
This tool creates a new session record in Supabase (a write operation), establishing a persistent session entry. It also loads context (a read operation), but the primary action is initializing/creating a new session. This is reversible and low blast radius — it simply starts a tracking session. No destructive, financial, or code execution behavior is indicated.
From the tool's definition Initialize a new memory session and load context. Call once at the start of each conversation.
Documented attack patterns abuse exactly the kind of access start_session gives an agent:
PolicyLayer is an MCP gateway — it sits between your AI agents and Claude Memory, and nothing reaches the server without passing your rules. This is the rule we recommend for start_session:
{
"version": "1",
"default": "deny",
"tools": {
"start_session": {
"limits": [
{
"counter": "start_session_rate",
"window": "minute",
"max": 30,
"scope": "grant"
}
]
}
}
}start_session stays usable, but capped — an agent stuck in a loop can't make hundreds of changes a minute. Everything else on the server is denied unless you say otherwise.
Free to start. No card required.
Initialize a new memory session and load context. Call once at the start of each conversation. It is categorised as a Write tool in the Claude Memory MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.
Register the Claude Memory MCP server in PolicyLayer and add a rule for start_session: 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 Memory. Nothing to install.
start_session is a Write tool with medium risk. Write tools should be rate-limited to prevent accidental bulk modifications.
Yes. Add a rate_limit block to the start_session 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 start_session. 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.
start_session is provided by the Claude Memory MCP server (jordanl61/claude-memory). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Start from Claude Memory, add the rest of your stack, and see everything your agents can call. Then put policy on all of it.
Free to start. No card required.
7 Claude Memory tools catalogued and risk-classified — across an index of 43,000+ MCP servers.