Read a session-scoped working-memory entry. Returns null if missing or expired.
AI agents call memory_get_working to retrieve information from Threadwork without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool retrieves session state data from working memory without creating, modifying, or deleting anything. It is a simple getter that queries existing session-scoped state. The 'Read' category applies to all retrieval operations regardless of data type or scope.
From the tool's definition Tool name 'memory_get_working' and description 'Read a session-scoped working-memory entry' explicitly indicate a read-only retrieval operation with no side effects. Returns data or null, no modifications possible.
Attacks that exploit this kind of access
Read a session-scoped working-memory entry. Returns null if missing or expired. It is categorised as a Read tool in the Threadwork MCP Server, which means it retrieves data without modifying state.
Register the Threadwork MCP server in PolicyLayer and add a rule for memory_get_working: 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 Threadwork. Nothing to install.
memory_get_working 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 memory_get_working 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_get_working. 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_get_working is provided by the Threadwork MCP server (tianqbu/threadwork). 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 →