Searches the user's stored facts and session history using hybrid semantic + keyword retrieval. Result content is capped to keep strict MCP clients under response limits; pass full_content=true only when the full matching row is required. Use whenever the user asks about anything that might be st...
AI agents call search_memory to retrieve information from UnClick without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
| Parameter | Type | Required | Description |
|---|---|---|---|
as_of | string | — | ISO 8601 timestamp for point-in-time queries (returns facts valid at that moment) |
query | string | Yes | Search query |
max_results | number | — | |
full_content | boolean | — | When true, returns full result content instead of 800-character previews. Use sparingly for strict clients. |
include_card | boolean | — | Phase 1 Wizard opt-in. When true, the response is { results, card } where card is a ConversationalCard summarising the matches for friendly chat surfaces. When |
Parameters from the server's own tool schema.
This tool purely retrieves and queries existing data from the user's memory store. It does not create, modify, delete, or execute external actions. The mention of capping results and optional full_content parameter indicates read-only access patterns. No destructive, financial, or executable operations are described. This is a straightforward Read operation typical of search/retrieval tools.
From the tool's definition Tool description states it 'Searches the user's stored facts and session history' with 'Result content is capped' and provides retrieval of stored data without modification or deletion.
Risk signalsAccepts freeform code/query input (query)
Attacks that exploit this kind of access
Searches the user's stored facts and session history using hybrid semantic + keyword retrieval. Result content is capped to keep strict MCP clients under response limits; pass full_content=true only when the full matching row is required. Use whenever the user asks about anything that might be stored: 'remember', 'recall', 'do you know', 'what did I say about', 'last time', 'context', 'profile', 'facts about me', 'who am I', 'my preferences', 'what have I told you', or when you need background on a topic before answering. Trigger even when the user doesn't explicitly say 'search' -- if the question involves past decisions, preferences, project details, or named people and tools, check memory first. Do NOT trigger for one-shot math, translations, definitions, or questions with no plausible stored context. Do NOT trigger if load_memory was just called and already returned the relevant context. It is categorised as a Read tool in the UnClick MCP Server, which means it retrieves data without modifying state.
search_memory accepts 5 parameters: as_of, query, max_results, full_content, include_card. Required: query. The full parameter table on this page comes from the server's own tool schema.
Register the UnClick MCP server in PolicyLayer and add a rule for search_memory: 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 UnClick. Nothing to install.
search_memory 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 search_memory 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 search_memory. 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.
search_memory is provided by the UnClick MCP server (@unclick/mcp-server). 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 →