Use this tool to classify a refund request into one of the 8 playbook cases (TH1 to TH8) and get back the recommended action, deduction and escalation flags. Common use-cases include: - Deciding whether to apply 0%, 10%, 20% or 40% deduction - Deciding whether to DECLINE (TH8) — the cycle was ful...
AI agents use classify_refund_case to commit financial operations through PageFly Refund MCP Server — usually the final step of a payment, billing, or trading workflow. A call moves real money.
classify_refund_case moves real money, and an autonomous agent will call it with the same confidence it calls a search tool. A misread instruction or an injected prompt is all it takes to drain an account or blow a budget.
Attacks that exploit this kind of access
Use this tool to classify a refund request into one of the 8 playbook cases (TH1 to TH8) and get back the recommended action, deduction and escalation flags. Common use-cases include: - Deciding whether to apply 0%, 10%, 20% or 40% deduction - Deciding whether to DECLINE (TH8) — the cycle was fully used or usage data (check_usage_data) proves active use, and the customer is not loyal/at-risk - Knowing whether the agent can self-decide or must escalate to Manager (Boo) / Shift Manager — and WHY (manager_reason) - Learning whether the store must first downgrade to Free and whether the bill must be Paid before the refund can be issued Pass the sensitive-case flags accurately: subscription_age_years (2+ = loyal), is_high_value, is_frustrated, bad_review_risk, already_left_bad_review and discount_commitment_claim ALL force Manager review and PREVENT an automatic decline — a bad review from a loyal customer costs far more than one cycle. Pass feature_issue / service_failure / is_trial_period / is_returning_customer for full-refund (0%) situations, and is_yearly_plan / customer_counters_deduction for the 10%. It is categorised as a Financial tool in the PageFly Refund MCP Server MCP Server, which means it involves financial transactions. Block by default and require explicit approval.
Register the PageFly Refund MCP Server MCP server in PolicyLayer and add a rule for classify_refund_case: 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 PageFly Refund MCP Server. Nothing to install.
classify_refund_case is a Financial 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 classify_refund_case 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 classify_refund_case. 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.
classify_refund_case is provided by the PageFly Refund MCP Server MCP server (nthieu002/refund-crisp-mcp). 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 →