Read or write the active working memory (scratchpad) for a user.
AI agents use memory_context to create or update resources in MemoryClaw — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your MemoryClaw environment.
Although the tool supports both read and write operations, the write capability to a user's active working memory (scratchpad) makes it a Write category tool. The severity is medium because: (1) modifications are reversible (scratchpad content can be overwritten or cleared), (2) the blast radius is limited to that user's working memory context, and (3) it does not delete data permanently or affect other systems.
From the tool's definition Tool description states 'Read or write the active working memory' — explicitly includes write capability. The dual read/write nature and access to active working memory elevates this beyond pure Read.
Documented attack patterns abuse exactly the kind of access memory_context gives an agent:
PolicyLayer is an MCP gateway — it sits between your AI agents and MemoryClaw, and nothing reaches the server without passing your rules. This is the rule we recommend for memory_context:
{
"version": "1",
"default": "deny",
"tools": {
"memory_context": {
"limits": [
{
"counter": "memory_context_rate",
"window": "minute",
"max": 30,
"scope": "grant"
}
]
}
}
}memory_context 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.
Read or write the active working memory (scratchpad) for a user. It is categorised as a Write tool in the MemoryClaw MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.
Register the MemoryClaw MCP server in PolicyLayer and add a rule for memory_context: 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 MemoryClaw. Nothing to install.
memory_context 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_context 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_context. 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_context is provided by the MemoryClaw MCP server (tostechbr/memoryclaw). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Start from MemoryClaw, 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.
6 MemoryClaw tools catalogued and risk-classified — across an index of 43,000+ MCP servers.