session_start
RARELY NEEDED — the SessionStart hook auto-loads context every session. Call only for an explicit deep re-load (full memories + checkpoints + decisions + health).
This record as markdown: /tools/20alexl-claude-engram/session-start.md
What session_start does on Claude Engram
AI agents call session_start to retrieve information from Claude Engram without modifying anything. It is typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
Why session_start is rated Low
The tool loads existing context and memory data into the current session. It reads stored information (memories, checkpoints, decisions, health status) without creating, modifying, or deleting data. The description emphasizes it is 'RARELY NEEDED' and is a 'deep re-load', confirming it is a retrieval operation. Low severity because misuse would at worst cause redundant data loading.
From the tool's definition 'auto-loads context every session' and 'full memories + checkpoints + decisions + health' — this is a read/load operation that retrieves stored session state
Risk signalsBulk/mass operation — affects multiple targets
Attacks that exploit this kind of access
The rule that runs session_start safely
PolicyLayer is an MCP gateway: it sits between your AI agents and Claude Engram, and checks every tool call against a rule you set before the call runs. Nothing changes on the server itself. For session_start, this is the rule to start with:
session_start is read-only, so it stays allowed. Everything else on the server is denied unless you say otherwise.
The button opens the PolicyLayer dashboard: create your workspace, connect Claude Engram, apply this rule, and every session_start call is checked against it from then on.
Questions about session_start
RARELY NEEDED — the SessionStart hook auto-loads context every session. Call only for an explicit deep re-load (full memories + checkpoints + decisions + health). It is categorised as a Read tool in the Claude Engram MCP Server, which means it retrieves data without modifying state.
Register the Claude Engram MCP server in PolicyLayer and add a rule for session_start: 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 Engram. Nothing to install.
session_start 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 session_start 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 session_start. 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.
session_start is provided by the Claude Engram MCP server (20alexl/claude-engram). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
More on Claude Engram, and thousands of servers like it.
This server
Across the catalogue