What it returns

Field	Meaning
days	Lookback window in days, counted back from now. Ignored when `start` is given.
start	Start of the range, ISO-8601 (e.g. '2026-08-01'). Overrides `days`.
end	End of the range, ISO-8601. Defaults to now.
exercise	Exercise name (e.g. 'Squat (Barbell)') or exercise template id.

Parameters

Name	Type	Description
`days`	`integer`	Lookback window in days, counted back from now. Ignored when `start` is given.
`start`	`string`	Start of the range, ISO-8601 (e.g. '2026-08-01'). Overrides `days`.
`end`	`string`	End of the range, ISO-8601. Defaults to now.
`exercise`	`string`	Exercise name (e.g. 'Squat (Barbell)') or exercise template id.

Pass either days or an explicit start/end pair. start wins if both are given.

Output

A markdown table rather than JSON — roughly a third of the tokens, and models reason over rows more reliably than nested objects.

Notes and gotchas

Requires a connected Hevy account. Connect it from your dashboard — no developer account needed.
Read-only: this tool cannot change anything in your account.

Call it directly

The endpoint is plain JSON-RPC, so you can test without any client:

curl

curl -s https://fitnesspeak.app/mcp \
  -H "Authorization: Bearer fp_live_YOUR_KEY_HERE" \
  -H "Content-Type: application/json" \
  -d '{
       "jsonrpc": "2.0",
       "id": 1,
       "method": "tools/call",
       "params": {
         "name": "hevy_exercise_history",
         "arguments": {
           "days": 7
         }
       }
     }'

Tool description the model sees

Every logged set of one exercise over time, one row per session: sets, top set, estimated 1RM, volume and average RPE. Call this when the user asks whether a lift is progressing, what their best or heaviest set was, or how much they were doing on a given exercise. `exercise` accepts a plain name ('bench press') or an exercise template id.

Descriptions say when to call a tool, not just what it does — that trigger wording is what models actually route on.

`hevy_exercise_history`