Synthex Forge — x402 Paid Agent Services

Recover Failed Workflow

recover_failed_workflow

Read-onlyIdempotent

The three answers a failed run needs are one decision. use it rather than recover_agent_workflow for failed workflow, not agent workflow; an object with 85 fields including action, canonical_request. $0.26 USDC.

Input Schema

Name	Required	Description
`workflow.step_log`	No	[POST /v1/autonomy/workflow-recovery]
`idempotency.history`	No	[POST /v1/operations/idempotency-check]
`idempotency.request`	No	[POST /v1/operations/idempotency-check]
`workflow.workflow_id`	No	[POST /v1/autonomy/workflow-recovery]
`idempotency.ignore_fields`	No	[POST /v1/operations/idempotency-check]
`idempotency.near_duplicate_threshold`	No	[POST /v1/operations/idempotency-check]

Output Schema

Name	Required	Description
`ok`	No	Whether every inner call succeeded.
`ts`	No	The `ts` field of the folded answer.
`done`	No	The `done` field of the folded answer.
`fail`	No	The `fail` field of the folded answer.
`skip`	No	The `skip` field of the folded answer.
`step`	No	The `step` field of the folded answer.
`todo`	No	The `todo` field of the folded answer.
`type`	No	The `type` field of the folded answer.
`entry`	No	The `entry` field of the folded answer.
`error`	No	The `error` field of the folded answer.
`field`	No	The `field` field of the folded answer.
`index`	No	The `index` field of the folded answer.
`state`	No	The `state` field of the folded answer.
`steps`	No	The `steps` field of the folded answer.
`action`	No	The `action` field of the folded answer.
`detail`	No	The `detail` field of the folded answer.
`queued`	No	The `queued` field of the folded answer.
`reason`	No	The `reason` field of the folded answer.
`status`	No	The `status` field of the folded answer.
`errored`	No	The `errored` field of the folded answer.
`failure`	No	The `failure` field of the folded answer.
`results`	No	One result per inner call.
`running`	No	The `running` field of the folded answer.
`started`	No	The `started` field of the folded answer.
`success`	No	The `success` field of the folded answer.
`summary`	No	One-line summary of the folded answer.
`complete`	No	The `complete` field of the folded answer.
`coverage`	No	The `coverage` field of the folded answer.
`problems`	No	The `problems` field of the folded answer.
`exception`	No	The `exception` field of the folded answer.
`executing`	No	The `executing` field of the folded answer.
`in_flight`	No	The `in_flight` field of the folded answer.
`operation`	No	The `operation` field of the folded answer.
`principal`	No	The `principal` field of the folded answer.
`succeeded`	No	The `succeeded` field of the folded answer.
`near_match`	No	The `near_match` field of the folded answer.
`similarity`	No	The `similarity` field of the folded answer.
`detected_by`	No	The `detected_by` field of the folded answer.
`differences`	No	The `differences` field of the folded answer.
`exact_match`	No	The `exact_match` field of the folded answer.
`fingerprint`	No	The `fingerprint` field of the folded answer.
`not_started`	No	The `not_started` field of the folded answer.
`occurrences`	No	The `occurrences` field of the folded answer.
`recorded_at`	No	The `recorded_at` field of the folded answer.
`resume_step`	No	The `resume_step` field of the folded answer.
`unambiguous`	No	The `unambiguous` field of the folded answer.
`workflow_id`	No	The `workflow_id` field of the folded answer.
`entry_sha256`	No	The `entry_sha256` field of the folded answer.
`failed_count`	No	The `failed_count` field of the folded answer.
`resume_point`	No	The `resume_point` field of the folded answer.
`this_request`	No	The `this_request` field of the folded answer.
`history_entry`	No	The `history_entry` field of the folded answer.
`matched_entry`	No	The `matched_entry` field of the folded answer.
`missing_steps`	No	The `missing_steps` field of the folded answer.
`pending_count`	No	The `pending_count` field of the folded answer.
`resume_safety`	No	The `resume_safety` field of the folded answer.
`skipped_count`	No	The `skipped_count` field of the folded answer.
`status_reason`	No	The `status_reason` field of the folded answer.
`ignored_fields`	No	The `ignored_fields` field of the folded answer.
`payload_sha256`	No	The `payload_sha256` field of the folded answer.
`request_sha256`	No	The `request_sha256` field of the folded answer.
`completed_count`	No	The `completed_count` field of the folded answer.
`completed_steps`	No	The `completed_steps` field of the folded answer.
`idempotency_key`	No	The `idempotency_key` field of the folded answer.
`near_duplicates`	No	The `near_duplicates` field of the folded answer.
`verdict_meaning`	No	The `verdict_meaning` field of the folded answer.
`duplicated_steps`	No	The `duplicated_steps` field of the folded answer.
`incomplete_steps`	No	The `incomplete_steps` field of the folded answer.
`canonical_request`	No	The `canonical_request` field of the folded answer.
`duplicate_details`	No	The `duplicate_details` field of the folded answer.
`history_available`	No	The `history_available` field of the folded answer.
`in_progress_count`	No	The `in_progress_count` field of the folded answer.
`last_completed_at`	No	The `last_completed_at` field of the folded answer.
`last_completed_step`	No	The `last_completed_step` field of the folded answer.
`resume_instructions`	No	The `resume_instructions` field of the folded answer.
`history_entry_sha256`	No	The `history_entry_sha256` field of the folded answer.
`last_completed_index`	No	The `last_completed_index` field of the folded answer.
`unknown_status_count`	No	The `unknown_status_count` field of the folded answer.
`unknown_status_steps`	No	The `unknown_status_steps` field of the folded answer.
`idempotency_key_reuse`	No	The `idempotency_key_reuse` field of the folded answer.
`idempotency_key_sha256`	No	The `idempotency_key_sha256` field of the folded answer.
`operation_scope_sha256`	No	The `operation_scope_sha256` field of the folded answer.
`idempotency_key_present`	No	The `idempotency_key_present` field of the folded answer.
`near_duplicate_threshold`	No	The `near_duplicate_threshold` field of the folded answer.
`history_entries_considered`	No	The `history_entries_considered` field of the folded answer.

Schema Changelog

Changes observed during successful MCP inspections.

Changed1 schema field changedOct 31, 2026

changedOutput schema / (root)

Previous value: -nullNew value: +{
+  "properties": {
+    "action": {
+      "description": "The `action` field of the folded answer.",
+      "type": "object"
+    },
+    "canonical_request": {
+      "description": "The `canonical_request` field of the folded answer.",
+      "type": "object"
+    },
+    "complete": {
+      "description": "The `complete` field of the folded answer.",
+      "type": "object"
+    },
+    "completed_count": {
+      "description": "The `completed_count` field of the folded answer.",
+      "type": "object"
+    },
+    "completed_steps": {
+      "description": "The `completed_steps` field of the folded answer.",
+      "type": "object"
+    },
+    "coverage": {
+      "description": "The `coverage` field of the folded answer.",
+      "type": "object"
+    },
+    "detail": {
+      "description": "The `detail` field of the folded answer.",
+      "type": "object"
+    },
+    "detected_by": {
+      "description": "The `detected_by` field of the folded answer.",
+      "type": "object"
+    },
+    "differences": {
+      "description": "The `differences` field of the folded answer.",
+      "type": "object"
+    },
+    "done": {
+      "description": "The `done` field of the folded answer.",
+      "type": "object"
+    },
+    "duplicate_details": {
+      "description": "The `duplicate_details` field of the folded answer.",
+      "type": "object"
+    },
+    "duplicated_steps": {
+      "description": "The `duplicated_steps` field of the folded answer.",
+      "type": "object"
+    },
+    "entry": {
+      "description": "The `entry` field of the folded answer.",
+      "type": "object"
+    },
+    "entry_sha256": {
+      "description": "The `entry_sha256` field of the folded answer.",
+      "type": "object"
+    },
+    "error": {
+      "description": "The `error` field of the folded answer.",
+      "type": "object"
+    },
+    "errored": {
+      "description": "The `errored` field of the folded answer.",
+      "type": "object"
+    },
+    "exact_match": {
+      "description": "The `exact_match` field of the folded answer.",
+      "type": "object"
+    },
+    "exception": {
+      "description": "The `exception` field of the folded answer.",
+      "type": "object"
+    },
+    "executing": {
+      "description": "The `executing` field of the folded answer.",
+      "type": "object"
+    },
+    "fail": {
+      "description": "The `fail` field of the folded answer.",
+      "type": "object"
+    },
+    "failed_count": {
+      "description": "The `failed_count` field of the folded answer.",
+      "type": "object"
+    },
+    "failure": {
+      "description": "The `failure` field of the folded answer.",
+      "type": "object"
+    },
+    "field": {
+      "description": "The `field` field of the folded answer.",
+      "type": "object"
+    },
+    "fingerprint": {
+      "description": "The `fingerprint` field of the folded answer.",
+      "type": "object"
+    },
+    "history_available": {
+      "description": "The `history_available` field of the folded answer.",
+      "type": "object"
+    },
+    "history_entries_considered": {
+      "description": "The `history_entries_considered` field of the folded answer.",
+      "type": "object"
+    },
+    "history_entry": {
+      "description": "The `history_entry` field of the folded answer.",
+      "type": "object"
+    },
+    "history_entry_sha256": {
+      "description": "The `history_entry_sha256` field of the folded answer.",
+      "type": "object"
+    },
+    "idempotency_key": {
+      "description": "The `idempotency_key` field of the folded answer.",
+      "type": "object"
+    },
+    "idempotency_key_present": {
+      "description": "The `idempotency_key_present` field of the folded answer.",
+      "type": "object"
+    },
+    "idempotency_key_reuse": {
+      "description": "The `idempotency_key_reuse` field of the folded answer.",
+      "type": "object"
+    },
+    "idempotency_key_sha256": {
+      "description": "The `idempotency_key_sha256` field of the folded answer.",
+      "type": "object"
+    },
+    "ignored_fields": {
+      "description": "The `ignored_fields` field of the folded answer.",
+      "type": "object"
+    },
+    "in_flight": {
+      "description": "The `in_flight` field of the folded answer.",
+      "type": "object"
+    },
+    "in_progress_count": {
+      "description": "The `in_progress_count` field of the folded answer.",
+      "type": "object"
+    },
+    "incomplete_steps": {
+      "description": "The `incomplete_steps` field of the folded answer.",
+      "type": "object"
+    },
+    "index": {
+      "description": "The `index` field of the folded answer.",
+      "type": "object"
+    },
+    "last_completed_at": {
+      "description": "The `last_completed_at` field of the folded answer.",
+      "type": "object"
+    },
+    "last_completed_index": {
+      "description": "The `last_completed_index` field of the folded answer.",
+      "type": "object"
+    },
+    "last_completed_step": {
+      "description": "The `last_completed_step` field of the folded answer.",
+      "type": "object"
+    },
+    "matched_entry": {
+      "description": "The `matched_entry` field of the folded answer.",
+      "type": "object"
+    },
+    "missing_steps": {
+      "description": "The `missing_steps` field of the folded answer.",
+      "type": "object"
+    },
+    "near_duplicate_threshold": {
+      "description": "The `near_duplicate_threshold` field of the folded answer.",
+      "type": "object"
+    },
+    "near_duplicates": {
+      "description": "The `near_duplicates` field of the folded answer.",
+      "type": "object"
+    },
+    "near_match": {
+      "description": "The `near_match` field of the folded answer.",
+      "type": "object"
+    },
+    "not_started": {
+      "description": "The `not_started` field of the folded answer.",
+      "type": "object"
+    },
+    "occurrences": {
+      "description": "The `occurrences` field of the folded answer.",
+      "type": "object"
+    },
+    "ok": {
+      "description": "Whether every inner call succeeded.",
+      "type": "boolean"
+    },
+    "operation": {
+      "description": "The `operation` field of the folded answer.",
+      "type": "object"
+    },
+    "operation_scope_sha256": {
+      "description": "The `operation_scope_sha256` field of the folded answer.",
+      "type": "object"
+    },
+    "payload_sha256": {
+      "description": "The `payload_sha256` field of the folded answer.",
+      "type": "object"
+    },
+    "pending_count": {
+      "description": "The `pending_count` field of the folded answer.",
+      "type": "object"
+    },
+    "principal": {
+      "description": "The `principal` field of the folded answer.",
+      "type": "object"
+    },
+    "problems": {
+      "description": "The `problems` field of the folded answer.",
+      "type": "object"
+    },
+    "queued": {
+      "description": "The `queued` field of the folded answer.",
+      "type": "object"
+    },
+    "reason": {
+      "description": "The `reason` field of the folded answer.",
+      "type": "object"
+    },
+    "recorded_at": {
+      "description": "The `recorded_at` field of the folded answer.",
+      "type": "object"
+    },
+    "request_sha256": {
+      "description": "The `request_sha256` field of the folded answer.",
+      "type": "object"
+    },
+    "results": {
+      "description": "One result per inner call.",
+      "items": {
+        "type": "object"
+      },
+      "type": "array"
+    },
+    "resume_instructions": {
+      "description": "The `resume_instructions` field of the folded answer.",
+      "type": "object"
+    },
+    "resume_point": {
+      "description": "The `resume_point` field of the folded answer.",
+      "type": "object"
+    },
+    "resume_safety": {
+      "description": "The `resume_safety` field of the folded answer.",
+      "type": "object"
+    },
+    "resume_step": {
+      "description": "The `resume_step` field of the folded answer.",
+      "type": "object"
+    },
+    "running": {
+      "description": "The `running` field of the folded answer.",
+      "type": "object"
+    },
+    "similarity": {
+      "description": "The `similarity` field of the folded answer.",
+      "type": "object"
+    },
+    "skip": {
+      "description": "The `skip` field of the folded answer.",
+      "type": "object"
+    },
+    "skipped_count": {
+      "description": "The `skipped_count` field of the folded answer.",
+      "type": "object"
+    },
+    "started": {
+      "description": "The `started` field of the folded answer.",
+      "type": "object"
+    },
+    "state": {
+      "description": "The `state` field of the folded answer.",
+      "type": "object"
+    },
+    "status": {
+      "description": "The `status` field of the folded answer.",
+      "type": "object"
+    },
+    "status_reason": {
+      "description": "The `status_reason` field of the folded answer.",
+      "type": "object"
+    },
+    "step": {
+      "description": "The `step` field of the folded answer.",
+      "type": "object"
+    },
+    "steps": {
+      "description": "The `steps` field of the folded answer.",
+      "type": "object"
+    },
+    "succeeded": {
+      "description": "The `succeeded` field of the folded answer.",
+      "type": "object"
+    },
+    "success": {
+      "description": "The `success` field of the folded answer.",
+      "type": "object"
+    },
+    "summary": {
+      "description": "One-line summary of the folded answer.",
+      "type": "string"
+    },
+    "this_request": {
+      "description": "The `this_request` field of the folded answer.",
+      "type": "object"
+    },
+    "todo": {
+      "description": "The `todo` field of the folded answer.",
+      "type": "object"
+    },
+    "ts": {
+      "description": "The `ts` field of the folded answer.",
+      "type": "object"
+    },
+    "type": {
+      "description": "The `type` field of the folded answer.",
+      "type": "object"
+    },
+    "unambiguous": {
+      "description": "The `unambiguous` field of the folded answer.",
+      "type": "object"
+    },
+    "unknown_status_count": {
+      "description": "The `unknown_status_count` field of the folded answer.",
+      "type": "object"
+    },
+    "unknown_status_steps": {
+      "description": "The `unknown_status_steps` field of the folded answer.",
+      "type": "object"
+    },
+    "verdict_meaning": {
+      "description": "The `verdict_meaning` field of the folded answer.",
+      "type": "object"
+    },
+    "workflow_id": {
+      "description": "The `workflow_id` field of the folded answer.",
+      "type": "object"
+    }
+  },
+  "type": "object"
+}

AddedOct 31, 2026

TDQS

B3.2/5.0

Behavior3/5

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

Annotations already declare readOnlyHint, idempotentHint, and non-destructive behavior, so the safety profile is covered. The description adds genuinely new context — that the result is a large decision object ('85 fields including action, canonical_request') and that each call costs $0.26 USDC — but it says nothing about permissions, failure modes, or when the decision is inconclusive. Notably, the name suggests an act of recovery while the description frames it as a decision object, which mildly conflicts with the read-only name but not with the annotations.

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?

It is short, and the sibling routing is front-loaded, which is good. But the sentence is a semicolon-spliced run-on that mixes an aphorism, a routing rule, an output shape, and a price with no clear hierarchy, and the opening line earns its place least.

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

Completeness3/5

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

An output schema exists, so return values needn't be enumerated, and the sibling routing plus cost note cover the main decision points. However, the description leaves ambiguous whether the tool merely produces a decision or performs the recovery, which is the single most important thing an agent needs to know before calling 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 100% and all six parameters carry endpoint-referencing descriptions, so the schema does the heavy lifting and a 3 is the baseline. The description adds no parameter meaning beyond that — inputs like step log, idempotency history, and near-duplicate threshold are left entirely to the schema.

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

Purpose3/5

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

The opening sentence ('The three answers a failed run needs are one decision') is cryptic and never states a clear verb+resource; the only concrete signal is the sibling contrast with `recover_agent_workflow` for 'failed workflow, not agent workflow', which implies a recovery-decision computation. An agent can infer the domain from the name and the sibling clause, but the description itself never plainly says what the tool returns or does.

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

Usage Guidelines4/5

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

It explicitly names the alternative (`recover_agent_workflow`) and the condition that selects this tool instead: 'failed workflow, not agent workflow'. That is a real routing rule, not implied. It stops short of stating any other exclusions or prerequisites.

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

All Synthex Forge — x402 Paid 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.