Check current context window usage and remaining capacity. Returns tokens used, remaining, percentage, and status (sufficient/low/critical).
AI agents call check_context_budget to retrieve information from Claude Session MCP without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool retrieves and returns contextual session information (tokens used, remaining capacity, percentage utilization, status flags) without creating, modifying, deleting, or executing any operations. It is a passive monitoring/diagnostic function analogous to checking system metrics.
From the tool's definition Tool name 'check_context_budget' and description 'Check current context window usage and remaining capacity.
Attacks that exploit this kind of access
Check current context window usage and remaining capacity. Returns tokens used, remaining, percentage, and status (sufficient/low/critical). It is categorised as a Read tool in the Claude Session MCP MCP Server, which means it retrieves data without modifying state.
Register the Claude Session MCP server in PolicyLayer and add a rule for check_context_budget: 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 Claude Session MCP. Nothing to install.
check_context_budget 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 check_context_budget 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 check_context_budget. 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.
check_context_budget is provided by the Claude Session MCP server (timevans/ccsession). 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 →