memory_store_scratch
AI agents use memory_store_scratch to create or update resources in Rekal — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your Rekal environment.
The name implies writing or persisting data to a scratch/temporary memory store. Given the server context (local SQLite-based memory system) and sibling tools (memory_build_context, memory_link, memory_prune, memory_delete), this tool most likely creates or updates a scratch memory record. Without a description, confidence is reduced, but 'store' strongly implies a Write operation.
From the tool's definition Tool name 'memory_store_scratch' suggests storing/writing scratch (temporary) memory data; description is empty and uninformative.
Documented attack patterns abuse exactly the kind of access memory_store_scratch gives an agent:
PolicyLayer is an MCP gateway — it sits between your AI agents and Rekal, and nothing reaches the server without passing your rules. This is the rule we recommend for memory_store_scratch:
{
"version": "1",
"default": "deny",
"tools": {
"memory_store_scratch": {
"limits": [
{
"counter": "memory_store_scratch_rate",
"window": "minute",
"max": 30,
"scope": "grant"
}
]
}
}
}memory_store_scratch 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.
memory_store_scratch. It is categorised as a Write tool in the Rekal MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.
Register the Rekal MCP server in PolicyLayer and add a rule for memory_store_scratch: 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 Rekal. Nothing to install.
memory_store_scratch 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 memory_store_scratch 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 memory_store_scratch. 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.
memory_store_scratch is provided by the Rekal MCP server (janbjorge/rekal). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Start from Rekal, 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.
21 Rekal tools catalogued and risk-classified — across an index of 43,000+ MCP servers.