NexRex

Draft a training plan

create_training_plan

Draft a goal-based training plan from recent status and a race goal. Use it when no catalog plan fits (recommend_training_plans reports no_good_match) or a coach wants a custom plan. The caller's own athlete_id comes from get_current_user; the caller need not be in a coach roster. Draft-only: writes a cached draft, never the athlete's calendar, and uses one of the caller's daily draft allowances when generation starts. Returns promptly with queued/generating status; get_plan_draft polls progress and ready content. An uncharged allowance_limited result gives retry_at for the same handle. Identical inputs on this connection reuse an active/ready job until consumed or expired; terminal failure permits a new job and a new charge. Returns a draft_id for draft edits and save_training_plan (single-use, expires in 24h, usable only by this account); content_preview holds the first weeks and next_cursor continues in get_plan_draft. Pass exactly one of athlete_id or group_id; group_id drafts a Group Plan Template (saved with save_training_plan save_mode='template_only') from distance, weeks and target_time; race_id and race_date are ignored for groups. workout_preference reflects the coach's choice: 'ai_generate' (default) creates workouts, 'prefer_existing' reuses the org's workout library where it can, 'strict_existing' uses only library workouts and returns status no_existing_workouts when none match.

Tool call template

This tool declares no required parameters. Add optional parameters from the input schema as needed.

{
  "id": 1,
  "jsonrpc": "2.0",
  "method": "tools/call",
  "params": {
    "arguments": {},
    "name": "create_training_plan"
  }
}

Input Schema

Name	Required	Description	Default
`level`	No
`weeks`	No
`race_id`	No
`distance`	No
`group_id`	No
`race_date`	No
`athlete_id`	No
`start_date`	No
`preferences`	No
`target_time`	No
`workout_preference`	No

Output Schema

Name	Required	Description
`goal`	No
`draft`	No
`error`	No
`level`	No
`weeks`	No
`status`	No
`message`	No
`privacy`	No	Present when the athlete's settings withhold data.
`draft_id`	No
`group_id`	No
`retry_at`	No	Epoch milliseconds; an uncharged allowance-limited job resumes on a poll at or after this instant.
`warnings`	No
`next_step`	No	What can be done next, for example after a failure.
`operation`	No
`plan_name`	No
`athlete_id`	No
`expires_at`	No
`group_name`	No
`next_steps`	No
`assumptions`	No
`athlete_pbs`	No
`goal_source`	No
`next_cursor`	No	Reads the next page; null on the last page.
`reason_code`	No
`total_weeks`	No
`member_count`	No
`retry_at_iso`	No	The epoch field beside it as a UTC instant; the local day can differ.
`subject_kind`	No
`user_content`	No	Fields in this result that people wrote.
`changed_cells`	No
`phase_outline`	No
`expires_at_iso`	No	The epoch field beside it as a UTC instant; the local day can differ.
`missing_inputs`	No
`weeks_returned`	No
`content_preview`	No
`generation_state`	No	queued, generating, ready, failed or access_revoked; only ready content is editable/saveable.
`truncated_fields`	No	Lists cut to fit one response: rows returned and total.
`allowance_charged`	No	Audit of the stable Redis reservation; retries of this job do not charge again.
`duration_in_weeks`	No
`generated_content`	No	Fields in this result that NexRex's AI wrote.
`weekly_mileage_km`	No
`workout_preference`	No
`preferences_applied`	No
`session_definitions`	No	Sessions that several rows of this result share by session_ref.
`content_block_fields`	No	Column names of every content_blocks row, nested blocks rows included.
`recommended_start_date`	No
`training_principle_org_id`	No
`training_principle_source`	No

Schema Changelog

Changes observed during successful MCP inspections.

Changed3 schema fields changedNov 14, 2026

changedOutput schema / $defs / Activity / properties / content_blocks / anyOf

Previous value: -[
-  {
-    "items": {
-      "items": {
-        "anyOf": [
-          {
-            "type": "string"
-          },
-          {
-            "type": "integer"
-          },
-          {
-            "type": "number"
-          },
-          {
-            "type": "boolean"
-          },
-          {
-            "items": {
-              "anyOf": [
-                {
-                  "type": "string"
-                },
-                {
-                  "type": "integer"
-                },
-                {
-                  "type": "number"
-                },
-                {
-                  "type": "boolean"
-                },
-                {
-                  "type": "null"
-                }
-              ]
-            },
-            "type": "array"
-          },
-          {
-            "$ref": "#/$defs/LocalizedText"
-          },
-          {
-            "type": "null"
-          }
-        ]
-      },
-      "type": "array"
-    },
-    "type": "array"
-  },
-  {
-    "type": "null"
-  }
-]New value: +[
+  {
+    "items": {
+      "items": {
+        "anyOf": [
+          {
+            "type": "string"
+          },
+          {
+            "type": "integer"
+          },
+          {
+            "type": "number"
+          },
+          {
+            "type": "boolean"
+          },
+          {
+            "items": {
+              "anyOf": [
+                {
+                  "type": "string"
+                },
+                {
+                  "type": "integer"
+                },
+                {
+                  "type": "number"
+                },
+                {
+                  "type": "boolean"
+                },
+                {
+                  "type": "null"
+                }
+              ]
+            },
+            "type": "array"
+          },
+          {
+            "$ref": "#/$defs/LocalizedText"
+          },
+          {
+            "items": {
+              "type": "array"
+            },
+            "type": "array"
+          },
+          {
+            "type": "null"
+          }
+        ]
+      },
+      "type": "array"
+    },
+    "type": "array"
+  },
+  {
+    "type": "null"
+  }
+]

changedOutput schema / $defs / Activity / properties / content_blocks / description

Previous value: -"Interval rows; each row's values follow content_block_fields."New value: +"Workout steps as rows whose values follow content_block_fields; an interval's blocks value is its nested steps as rows in the same columns."

changedOutput schema / properties / content_block_fields / description

Previous value: -"Column names of every content_blocks row."New value: +"Column names of every content_blocks row, nested blocks rows included."

First observedNov 12, 2026

TDQS

A4.7/5.0

Behavior5/5

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

Going well beyond the readOnly/destructive/idempotent hints, it discloses that this is draft-only, writes a cached draft rather than the athlete's calendar, consumes a daily draft allowance, returns queued/generating promptly, and that allowance_limited is uncharged with a retry_at. It even explains job-reuse behavior for identical inputs on the same connection and when a new job and charge are permitted after terminal failure.

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

Conciseness4/5

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

The description is front-loaded with purpose, then eligibility, then behavior, then parameters, and nearly every sentence adds a distinct fact. It is a single dense block, however, and the allowance/retry and draft-lifecycle details are packed together in a way that takes effort to parse.

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

Completeness4/5

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

For a non-destructive but stateful generation tool, it covers side effects, cost, lifecycle, polling handoff to get_plan_draft, and the draft_id handoff to save_training_plan. The remaining gap is the semantics of the individual-mode planning inputs (level, start_date, preferences) and the fact that no parameter is marked required despite an exclusive-or rule.

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

Parameters4/5

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

With 0% schema description coverage across 11 parameters, the description carries the load and does so for the critical ones: exactly one of athlete_id or group_id, group mode describing a Group Plan Template using distance, weeks and target_time, race_id/race_date ignored for groups, and the three-valued workout_preference with its no_existing_workouts outcome. It leaves level, start_date, preferences and the general meaning of weeks/target_time in individual mode undocumented, so it is strong but not exhaustive.

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?

The opening sentence names a specific verb and artifact: 'Draft a goal-based training plan from recent status and a race goal.' It also implicitly separates the tool from the catalog-based sibling by naming recommend_training_plans and save_training_plan, so an agent can tell what this tool produces versus what those do.

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?

It gives an explicit trigger condition ('when no catalog plan fits (recommend_training_plans reports no_good_match) or a coach wants a custom plan') and names the alternative tool by name. It also states the caller need not be in a coach roster, which removes a common precondition guess.

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

All NexRex 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?