agent 1 credit · ~$0.015 <5ms
Tool-Argument Firewall
POST /v1/agent/tool-args A generic business-rules gate for any function/tool call an agent is about to make. Declare a schema (types, required, enums, ranges, patterns, units) plus cross-field rules, and it returns PASS/FLAG/BLOCK with a precise list of violations before the call executes — catching the class of bug where an agent passes dollars where cents are expected, a null into a required field, or a value outside its allowed range.
Capabilities
- ✓ Type, required, enum, range, length, and regex checks
- ✓ Null-safety and unknown-argument detection
- ✓ Unit-coercion heuristics (dollars vs cents, the Stripe ×100 bug)
- ✓ Cross-field rules (e.g. min <= max) with severity-based verdicts
- ✓ Deterministic and offline, with a SHA-256 certificate
Use cases
- → Gate any agent tool/function call touching money or quantities
- → Validate arguments before a destructive or irreversible action
- → Enforce business policy on LLM-generated parameters
- → A cheap, universal pre-execution guardrail
example
curl -X POST https://api.agent-toolbox.ai/v1/agent/tool-args \
-H "Content-Type: application/json" \
-d '{"toolName": "transfer", "args": {"amount": 99.99, "currency": "USD"}, "schema": {"amount": {"type": "number", "unit": "cents"}}}' response shape
{
"verdict": "PASS" | "FLAG" | "BLOCK",
// tool-specific findings…
"certificate": "sha256:..."
}