DropEngine x402 Agent Services

preflight_action

Agent Action Gate ($0.02 USDC flat; stable MCP tool name preflight_action). Use before an AI agent performs an external action. Evaluates the required safety, trust, spending and transaction checks and returns ALLOW, REVIEW, REQUIRE_APPROVAL or DENY. Use before opening an untrusted URL, installing a dependency, calling an unfamiliar MCP/tool, spending funds, submitting a transaction, or consuming untrusted external content. Do not use for pure local deterministic computation or when no external action/risk exists. Does not assess seller reputation. Missing or skipped relevant checks cannot produce ALLOW. REQUIRE_APPROVAL is a recommendation only; use /api/financial/authorize-spend separately for actual spend-policy authorization and owner approval. It does not sign, route payment, or broadcast. No private keys, seed phrases, or secrets are accepted. Typical latency depends on selected checks/providers; there is no uniform SLA. Recommended workflows: agent-safety-gate, safe-transaction-preflight, tool-discovery-trust.

Agent entry points: risky external action → use Agent Action Gate when a combined action-specific safety assessment is needed; unclear capability/provider → use free /api/discover; constrained comparison between callable alternatives → use paid route_tool; known single capability → call the direct tool; multi-step objective → read the matching workflow recipe and call only relevant steps. Workflows are stateless recipes, not executors; calls are paid separately and are not bundle charged. Agent Action Gate does not approve payment, sign or broadcast transactions, evaluate seller reputation, replace transaction simulation, replace route_tool, or replace human approval.

Input Schema

TableJSON Schema

Name	Required	Description	Default
`intent`	Yes
`target`	No
`context`	No
`action_type`	Yes
`expected_cost`	No

Output Schema

TableJSON Schema

Name	Required	Description	Default
`meta`	No
`next`	No
`risk`	Yes
`error`	No
`checks`	Yes
`intent`	Yes
`result`	No
`status`	No
`payment`	No
`reasons`	Yes
`success`	Yes
`agent_ux`	No
`decision`	Yes
`retryable`	Yes
`checks_run`	Yes
`confidence`	Yes
`latency_ms`	Yes
`action_type`	Yes
`skip_reason`	Yes
`failed_checks`	Yes
`decision_scope`	Yes
`estimated_cost`	Yes
`next_best_tool`	Yes
`skipped_checks`	Yes
`evidence_status`	Yes
`required_checks`	Yes
`approval_handoff`	Yes
`attempted_checks`	Yes
`completed_checks`	Yes
`recommended_action`	Yes
`confidence_semantics`	Yes
`signing_or_broadcast_performed`	Yes

Schema Changelog

Changes observed during successful MCP inspections.

Changed13 schema fields changedOct 8, 2026

addedOutput schema / properties / agent_ux

Added value: +{
+  "additionalProperties": false,
+  "properties": {
+    "error": {
+      "anyOf": [
+        {},
+        {
+          "type": "null"
+        }
+      ]
+    },
+    "meta": {
+      "additionalProperties": {},
+      "propertyNames": {
+        "type": "string"
+      },
+      "type": "object"
+    },
+    "next": {
+      "items": {},
+      "type": "array"
+    },
+    "payment": {
+      "additionalProperties": false,
+      "properties": {
+        "execution_status": {
+          "enum": [
+            "not_started",
+            "succeeded",
+            "failed",
+            "unknown"
+          ],
+          "type": "string"
+        },
+        "status": {
+          "enum": [
+            "not_required",
+            "required",
+            "settled",
+            "failed",
+            "unknown"
+          ],
+          "type": "string"
+        }
+      },
+      "required": [
+        "status",
+        "execution_status"
+      ],
+      "type": "object"
+    },
+    "result": {},
+    "status": {
+      "enum": [
+        "succeeded",
+        "failed",
+        "partial",
+        "outcome_unknown"
+      ],
+      "type": "string"
+    }
+  },
+  "required": [
+    "status",
+    "result",
+    "error",
+    "meta",
+    "payment",
+    "next"
+  ],
+  "type": "object"
+}

addedOutput schema / properties / approval_handoff

Added value: +{
+  "additionalProperties": false,
+  "properties": {
+    "note": {
+      "type": "string"
+    },
+    "request_id": {
+      "type": "null"
+    },
+    "status": {
+      "enum": [
+        "not_requested",
+        "not_created_by_gate"
+      ],
+      "type": "string"
+    }
+  },
+  "required": [
+    "status",
+    "request_id",
+    "note"
+  ],
+  "type": "object"
+}

addedOutput schema / properties / attempted_checks

Added value: +{
+  "items": {
+    "type": "string"
+  },
+  "type": "array"
+}

addedOutput schema / properties / confidence_semantics

Added value: +{
+  "const": "Heuristic evidence sufficiency/risk signal; not a probability or guarantee.",
+  "type": "string"
+}

addedOutput schema / properties / decision_scope

Added value: +{
+  "additionalProperties": false,
+  "properties": {
+    "assesses": {
+      "items": {
+        "type": "string"
+      },
+      "type": "array"
+    },
+    "does_not": {
+      "items": {
+        "type": "string"
+      },
+      "type": "array"
+    },
+    "question": {
+      "const": "Should my agent perform this action?",
+      "type": "string"
+    }
+  },
+  "required": [
+    "question",
+    "assesses",
+    "does_not"
+  ],
+  "type": "object"
+}

addedOutput schema / properties / error

Added value: +{
+  "anyOf": [
+    {},
+    {
+      "type": "null"
+    }
+  ]
+}

addedOutput schema / properties / evidence_status

Added value: +{
+  "enum": [
+    "incomplete",
+    "rule_evaluable"
+  ],
+  "type": "string"
+}

addedOutput schema / properties / meta

Added value: +{
+  "additionalProperties": {},
+  "propertyNames": {
+    "type": "string"
+  },
+  "type": "object"
+}

addedOutput schema / properties / next

Added value: +{
+  "items": {},
+  "type": "array"
+}

addedOutput schema / properties / payment

Added value: +{
+  "additionalProperties": false,
+  "properties": {
+    "execution_status": {
+      "enum": [
+        "not_started",
+        "succeeded",
+        "failed",
+        "unknown"
+      ],
+      "type": "string"
+    },
+    "status": {
+      "enum": [
+        "not_required",
+        "required",
+        "settled",
+        "failed",
+        "unknown"
+      ],
+      "type": "string"
+    }
+  },
+  "required": [
+    "status",
+    "execution_status"
+  ],
+  "type": "object"
+}

addedOutput schema / properties / result
```
Added value: +{}
```

addedOutput schema / properties / status

Added value: +{
+  "enum": [
+    "succeeded",
+    "failed",
+    "partial",
+    "outcome_unknown"
+  ],
+  "type": "string"
+}

changedOutput schema / required

Previous value: -[
-  "success",
-  "decision",
-  "risk",
-  "confidence",
-  "required_checks",
-  "completed_checks",
-  "checks_run",
-  "failed_checks",
-  "skipped_checks",
-  "skip_reason",
-  "reasons",
-  "retryable",
-  "estimated_cost",
-  "recommended_action",
-  "next_best_tool",
-  "latency_ms",
-  "signing_or_broadcast_performed",
-  "checks",
-  "action_type",
-  "intent"
-]New value: +[
+  "success",
+  "decision",
+  "risk",
+  "confidence",
+  "confidence_semantics",
+  "evidence_status",
+  "required_checks",
+  "attempted_checks",
+  "completed_checks",
+  "checks_run",
+  "failed_checks",
+  "skipped_checks",
+  "skip_reason",
+  "reasons",
+  "retryable",
+  "estimated_cost",
+  "recommended_action",
+  "next_best_tool",
+  "latency_ms",
+  "signing_or_broadcast_performed",
+  "checks",
+  "decision_scope",
+  "approval_handoff",
+  "action_type",
+  "intent"
+]

AddedOct 8, 2026

TDQS

A4.5/5.0

Behavior5/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

With no annotations, the description carries the full burden and delivers: flat $0.02 USDC pricing, that missing/skipped checks cannot yield ALLOW, that REQUIRE_APPROVAL is a recommendation only, that it does not sign/route/broadcast payments, that no keys or secrets are accepted, and that latency has no uniform SLA. It also clarifies workflows are stateless recipes not bundle-charged executors.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness3/5

Is the description appropriately sized, front-loaded, and free of redundancy?

Front-loaded well (name, price, use case), but the text is dense and contains genuine repetition — 'It does not sign, route payment, or broadcast' restates the earlier 'does not approve payment, sign or broadcast transactions', and the non-replacement list is duplicated by 'Agent Action Gate does not... replace human approval'.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness5/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

An output schema exists, so return-value explanation is unnecessary. For a paid gate with nested params and enumerated decisions, the description supplies pricing, decision semantics, exclusions, sibling routing, and workflow guidance — everything needed to decide whether and how to invoke it.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters3/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema description coverage is 0% with five top-level params including nested target/context/expected_cost objects. The action-type list implicitly maps inputs to fields (URL→open_url, package→install_package, transaction→transaction, content→use_external_content), which adds real value, but intent, risk_tolerance, agent_id, purpose, expected_cost subfields and extract_fields remain undocumented in both schema and description.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

States a specific verb and resource ('Evaluates the required safety, trust, spending and transaction checks') and names the exact output vocabulary (ALLOW, REVIEW, REQUIRE_APPROVAL, DENY). It distinguishes itself from siblings by explicitly pointing to /api/discover, route_tool, and direct tools as separate paths, so an agent can tell it apart without opening a schema.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines5/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

Gives an explicit when-to-use list (untrusted URL, dependency install, unfamiliar MCP/tool, spending funds, submitting a transaction, consuming untrusted content), an explicit when-not ('pure local deterministic computation... no external action/risk'), and routing to alternatives by scenario. This leaves almost nothing to inference.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

All DropEngine x402 Agent Services tools Inspect

Try in Browser

One endpoint for every MCP toolRayrun runs every MCP server behind one URL and holds the credentials. Each call is decided per client, tool, and argument, then recorded.ray.runAd

Glama MCP Gateway

Add one secure layer between your agents and this server.

Resources

Need Help?