Skip to main content
Glama

Add an exercise

add_exercise

Adds an exercise to the catalog when the athlete wants one that list_exercises doesn't have. Write the name in sentence case, like the catalog's ("Single-leg calf raise", "KB sumo deadlift"): a capital on the first word only, plus abbreviations and people's names. The id is made from the name and must not clash with an existing one. kind is required: back (shown as Foundation), strength or mobility (or other), the block kind it goes in; set_today_session only accepts it in a block of that kind. For a mobility exercise also give style (dynamic, static or release; see list_exercises). Give sides when it is clear: together (both sides at once) or each (one side per set); start.bothSides must match it, and so must every later prescription. painAreas must name at least one body area it stresses. start is required: a conservative first-time prescription to begin from. It must stay within at most 3 sets, 15 reps or 120 s per set, at least 30 s rest, and weightKg only for a loaded exercise, at most 16 kg; anything above is rejected with every problem listed. Alternatives must be ids from list_exercises. guide is required: the how-to the athlete reads before trying it, written for someone who has never seen the exercise: setup (equipment and starting position), steps (2–6, one action each, in order), feel (where they should feel it) and avoid (the one common mistake). Plain words, no medical claims. Also give a one-line instructions for the set screen. videoUrl is optional (https); the app offers a video search when there is none. The exercise is saved as not yet reviewed by the athlete; once added, its id works in set_today_session.

Tool call template

Replace each placeholder with a value matching the input schema. Only required parameters are included; add optional parameters as needed.

{
  "id": 1,
  "jsonrpc": "2.0",
  "method": "tools/call",
  "params": {
    "arguments": {
      "name": "<name>",
      "kind": "<kind>",
      "equipment": "<equipment>",
      "loaded": "<loaded>",
      "painAreas": "<painAreas>",
      "start": "<start>"
    },
    "name": "add_exercise"
  }
}

Input Schema

NameRequiredDescriptionDefault
kindYes
nameYes
guideNoguide is required: { setup?, steps (2–6 short steps), feel?, avoid? }, written for someone who has never seen the exercise
sidesNo
startYes
styleNo
loadedYes
videoUrlNo
equipmentYes
painAreasYes
alternativesNo
instructionsNo

Schema Changelog

Changes observed during successful MCP inspections.

No schema history has been recorded yet.

TDQS

A4.8/5.0
Behavior5/5

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

Annotations only declare the mutation/safety profile; the description adds substantial behavior beyond them: the strict creation-time validation envelope (≤3 sets, ≤15 reps or ≤120 s, ≥30 s rest, ≤16 kg only when loaded), the fact that violations are rejected with every problem listed, and that the created exercise is saved as not yet reviewed. It also discloses id-derivation and clash behavior, plus the video-search fallback when videoUrl is absent.

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?

It is a single dense block, but nearly every clause carries a constraint an agent could otherwise get wrong, and the purpose and sibling routing are front-loaded. Minor cost: constraint clauses are chained with semicolons rather than separated, making it harder to scan than it needs to be for a 12-parameter tool.

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?

For a 12-parameter, nested, non-read-only tool with no output schema, the definition covers naming conventions, required fields, numeric limits, cross-field consistency (sides/bothSides), rejection behavior, and post-creation state. The only unstated element is the exact return payload, which is low-impact given it explains how the generated id is used afterward.

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 schema description coverage at 8%, the description carries the burden and mostly delivers: kind's meaning and Foundation display, style only for mobility, sides plus the requirement that start.bothSides and every later prescription match, painAreas needing at least one area, the guide's four sections with step counts, instructions, videoUrl scheme, and alternatives being list_exercises ids. Gaps remain on equipment, loaded, cue, and the distinction between the alternatives keys (noKit vs gentle), which the schema does not document either.

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 ("Adds an exercise to the catalog") and immediately distinguishes it from list_exercises by naming the exact condition that selects this tool instead. An agent can route correctly without opening the 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 rule ("when the athlete wants one that list_exercises doesn't have"), points to list_exercises for alternatives ids, and explains the downstream consequence that the new id works in set_today_session. It also notes the kind must match the block that set_today_session will accept, which is a real usage constraint.

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

Try in Browser

Glama MCP Gateway

Add one secure layer between your agents and this server.

Resources