Atlas — ask Atlas a question and get an answer (POST ${ATLAS_URL}/v1/ask). Atlas auto-classifies the prompt, routes to the best model, and returns { answer, model_used, task_type, latency_ms, tokens }. SINGLE-TURN and STATELESS by design — there is no server-side conversation memory. For multi-tu...
AI agents call atlas_ask to retrieve information from Celiums Memory without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
atlas_ask is a query/retrieval operation that takes a question as input and returns structured answers from the knowledge base. It performs no side effects, creates no persistent changes, executes no external code or commands, and does not modify or delete data. The explicit mention of being 'stateless' and lacking 'server-side conversation memory' confirms this is a read-only retrieval function.
From the tool's definition Tool description states it 'ask Atlas a question and get an answer' and is 'SINGLE-TURN and STATELESS by design — there is no server-side conversation memory.' The tool retrieves and returns information (answer, model_used, task_type, latency_ms, tokens) with…
Documented attack patterns abuse exactly the kind of access atlas_ask gives an agent:
PolicyLayer is an MCP gateway — it sits between your AI agents and Celiums Memory, and nothing reaches the server without passing your rules. This is the rule we recommend for atlas_ask:
{
"version": "1",
"default": "deny",
"tools": {
"atlas_ask": {}
}
}atlas_ask is read-only, so it stays allowed — but everything else on the server is denied unless you say otherwise.
Free to start. No card required.
Atlas — ask Atlas a question and get an answer (POST ${ATLAS_URL}/v1/ask). Atlas auto-classifies the prompt, routes to the best model, and returns { answer, model_used, task_type, latency_ms, tokens }. SINGLE-TURN and STATELESS by design — there is no server-side conversation memory. For multi-turn, use atlas_chat with an explicit \. It is categorised as a Read tool in the Celiums Memory MCP Server, which means it retrieves data without modifying state.
Register the Celiums Memory MCP server in PolicyLayer and add a rule for atlas_ask: 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 Celiums Memory. Nothing to install.
atlas_ask 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 atlas_ask 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 atlas_ask. 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.
atlas_ask is provided by the Celiums Memory MCP server (terrizoaguimor/celiums-memory). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Start from Celiums Memory, 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.
62 Celiums Memory tools catalogued and risk-classified — across an index of 43,000+ MCP servers.