Remove a context cache. Frees resources and stops billing for that cache.
AI agents call context_evict to permanently remove resources in Mnemo — typically in cleanup and lifecycle workflows. It does its job in a single call, and there is no undo.
This tool irreversibly deletes cached context data. While the data can theoretically be reloaded via context_load, the eviction itself is a destructive action that removes the resource. In the context of an AI agent, misuse could result in loss of critical cached information needed for subsequent operations, disruption of workflows, or unintended resource cleanup.
From the tool's definition Tool name 'context_evict' combined with description 'Remove a context cache' indicates deletion. The phrase 'Frees resources and stops billing for that cache' confirms the action is irreversible—once evicted, the cached data (potentially large codebases,…
Documented attack patterns abuse exactly the kind of access context_evict gives an agent:
PolicyLayer is an MCP gateway — it sits between your AI agents and Mnemo, and nothing reaches the server without passing your rules. This is the rule we recommend for context_evict:
{
"version": "1",
"default": "deny",
"hide": [
"context_evict"
]
}context_evict disappears from the agent's tool list entirely, and any attempt to call it is denied. The rest of the server keeps working.
Free to start. No card required.
Remove a context cache. Frees resources and stops billing for that cache. It is categorised as a Destructive tool in the Mnemo MCP Server, which means it can permanently delete or destroy data. Block by default and require explicit approval.
Register the Mnemo MCP server in PolicyLayer and add a rule for context_evict: 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 Mnemo. Nothing to install.
context_evict is a Destructive tool with critical risk. Critical-risk tools should be blocked by default and only enabled with explicit human approval.
Yes. Add a rate_limit block to the context_evict 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 context_evict. 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.
context_evict is provided by the Mnemo MCP server (logos-flux/mnemo). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Start from Mnemo, 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 Mnemo tools catalogued and risk-classified — across an index of 43,000+ MCP servers.