# io.github.dropflyai/fitfly-engine

> Deterministic fitness coaching engine: adaptive programs, progression math, readiness autoregulation

- name: io.github.dropflyai/fitfly-engine
- version: 0.2.0
- connection class: R1 (autonomous)
- trust: 73/100 flags: no-repository
- quality: 79/100 (good)
- owner: not claimed
- descriptor: https://api.protogrid.dev/v1/servers/io.github.dropflyai%2Ffitfly-engine
- tools: https://api.protogrid.dev/v1/servers/io.github.dropflyai%2Ffitfly-engine/tools

**An agent can connect on its own if it already holds `FITFLY_ENGINE_TOKEN`.** Remote endpoint behind a static API key or header. The registry never sees the value; the agent fills the placeholder.

## Connection (mcpServers)

```json
{
  "mcpServers": {
    "fitfly-engine": {
      "type": "http",
      "url": "https://workout-engine-production-5dd7.up.railway.app/mcp",
      "headers": {
        "Authorization": "Bearer ${FITFLY_ENGINE_TOKEN}"
      }
    }
  }
}
```

Placeholders to fill: `${FITFLY_ENGINE_TOKEN}`.

## Trust

- hygiene: 80
- liveness: 100
- freshness: 65
- provenance: 45
- drivers: +website +reachable +uptime-100% ~updated-111d-ago -no-repository
- Derived from observable signals (official registry feed and our own credential-free probes); no code audit performed.

## Quality 79/100 (good)

### protocol: 75 (weight 25)

- warn `protocol.modern`: newest supported version is 2025-11-25; 2026-07-28 not supported, older versions are deprecated until 2027-07-28
- n/a `protocol.stateless`: only applies to 2026-07-28 servers
- pass `protocol.transport`: streamable HTTP
- n/a `protocol.list_ttl`: only applies to 2026-07-28 servers

### authorization: n/a (weight 15)

- n/a `auth.prm`, `auth.as_metadata`, `auth.cimd`: no remote uses OAuth
- n/a `auth.secret_in_url`: no templated URL

### tool hygiene: 71 (weight 30)

- pass `tools.descriptions`: every tool has a description
- pass `tools.description_length`: descriptions are concise
- pass `tools.schemas`: every tool has an object input schema
- fail `tools.annotations`: no tool declares readOnlyHint or destructiveHint, so clients cannot tell safe calls from risky ones
- fail `tools.directory_hints`: no tool declares all four of readOnlyHint, destructiveHint, idempotentHint and openWorldHint; missing: list_swap_options (readOnlyHint, destructiveHint, idempotentHint, openWorldHint); nutrition_targets (readOnlyHint, destructiveHint, idempotentHint, openWorldHint); regenerate_program (readOnlyHint, destructiveHint, idempotentHint, openWorldHint); …
- pass `tools.token_cost`: about 1,125 tokens to load every tool
- pass `tools.api_dump`: tools are not a one-to-one API dump

### stability: 100 (weight 15)

- pass `stability.changes`: no tool changes in 30 days
- pass `stability.rug_pull`: no tool changed its meaning under the same name

### dependencies: n/a (weight 15)

- n/a `deps.known_vulns`, `deps.mcp_sdk_version`, `deps.resolvable`: no npm or PyPI package

- tools: 10, about 1,125 tokens to load them all
- tool set hash: 80c6e12dbbbc15fe2888e8561e43d2c2
- badge: [![protogrid quality](https://protogrid.dev/badge/io.github.dropflyai/fitfly-engine.svg)](https://protogrid.dev/servers/io.github.dropflyai/fitfly-engine)
- Checks run on what our credential-free, read-only probes observe; tools are never called and no code audit is performed.

### Tool changes

_No tool definition changes recorded._

## Remotes

- https://workout-engine-production-5dd7.up.railway.app/mcp (streamable-http, auth api_key ${FITFLY_ENGINE_TOKEN}, reachable true, uptime30d 1)

## Packages

_none_

## Tools (10)

- `changelog`: The 'what changed & why' feed — the engine's coaching decisions (volume bumps, deloads, swaps) in Coach Fly's voice. Use to explain to the athlete why their plan evolved.
- `get_program`: The athlete's current multi-day program (render-ready).
- `get_todays_session`: Today's workout: exercises, sets/reps, per-exercise load guidance and concrete target weights, plus nutrition targets and cycle status. Auto-applies the user's readiness check-in (autoregulation) if they've done one today. Call this to see what the athlete should do now.
- `list_swap_options`: Valid alternatives for an exercise — same muscle group, filtered to the athlete's equipment/level/injuries. Use before swap_exercise.
- `log_workout`: Record what the athlete actually lifted. The engine runs progression (double-progression, e1RM), detects PRs, and returns per-exercise outcomes (established/progressed/held/deloaded) with next-time targets. Weights are in kg.
- `meal_plan`: A day of meals hitting the athlete's targets, USDA truth-checked (flags feasible=false with a note when macros are approximate). Slow (~15s AI call).
- `nutrition_targets`: The athlete's daily calorie + macro targets, computed from body stats, activity, and goal.
- `readiness_checkin`: Record how the athlete feels today (energy, sleep, soreness — each 1–5). The engine scores readiness and, on the next get_todays_session, reshapes the workout accordingly (reduce volume / reduce intensity / recovery day). Returns the readiness score + modifier.
- `regenerate_program`: Rebuild the whole program from the stored profile (~20s AI generation). Use after a profile change or a plateau.
- `swap_exercise`: Replace an exercise with a valid same-area alternative. scope 'today' = just this session; scope 'program' = going forward. The new lift starts fresh (day-one calibration).