# leo-running-coach

Coach Leo is an AI running coach that syncs your real training data from Strava, remembers every conversation, tracks your body signals and shoes, and adapts your training to who you actually are as …

## Quick Start

```bash
# Connect this server (installs CLI if needed)
npx -y @smithery/cli@latest mcp add coachleo/running-coach

# Browse available tools
npx -y @smithery/cli@latest tool list coachleo/running-coach

# Get full schema for a tool
npx -y @smithery/cli@latest tool get coachleo/running-coach get_coaching_briefing

# Call a tool
npx -y @smithery/cli@latest tool call coachleo/running-coach get_coaching_briefing '{}'
```

## Direct MCP Connection

Endpoint: `https://leo--running-coach.run.tools`

## Tools (12)

- `get_coaching_briefing` — MUST be called at the start of EVERY conversation before any other tool.
- `manage_athlete` — Manage the runner's profile and running character. Use when the runner
- `manage_activities` — Access and annotate the runner's training activities. Use when the runner
- `manage_training` — Manage training plans and sessions.
- `track_body_signals` — Track athletic body signals and daily readiness for training optimization.
- `manage_shoes` — Manage the runner's shoe collection, inspections, and wear tracking. Use when
- `manage_races` — Manage the runner's races from discovery to post-race report. When a runner
- `search_knowledge` — Access and enrich the scientific knowledge base. CRITICAL: you MUST search
- `coaching_memory` — Persistent coaching memory across sessions. ALWAYS log meaningful coaching
- `estimate_vo2max` — Estimate VO2max using 3 scientific methods (ACSM Running Equation, Jack Daniels VDOT,
- `analyze_gpx` — Parse and analyze a GPX file: elevation profile, segments (climb/descent/flat),
- `get_skill_workflow` — Load a detailed coaching workflow by its ID. Call this when you detect a

```bash
# Get full input/output schema for a tool
npx -y @smithery/cli@latest tool get coachleo/running-coach <tool-name>
```

## Prompts (11)

- `onboarding` — Guides a new runner through initial setup when profile is empty or incomplete. Use when get_coaching_briefing() returns…
- `post-activity-analysis` — Analyzes completed running and trail activities, compares planned vs actual performance, and records athlete feedback. …
- `body-signal-analysis` — Analyzes training-related body signals: tightness, soreness, fatigue patterns using scientific knowledge and training h…
- `overtraining-monitor` — Monitors overtraining risk from training load spikes, recurring discomfort patterns, and recovery metrics with graduate…
- `weekly-review` — Produces comprehensive weekly training review with volume analysis, load trends, plan adherence, readiness patterns, an…
- `adaptive-training-plan` — Creates and adapts personalized training plans based on scientific principles, athlete profile, and race goals. Use whe…
- `race-planning` — Manages full race lifecycle from discovery through registration, preparation, taper, race-day strategy, to post-race de…
- `deep-research` — Enriches the scientific knowledge base by researching topics where gaps have been detected. Use when runner requests kn…
- `gpx-pattern-analysis` — Analyzes GPX route files for elevation profile, terrain segments, and diversity scoring. Use when runner shares a GPX f…
- `shoe-wear-analysis` — Analyzes running shoe wear from photos, scoring zones and detecting running form patterns. Use when runner shares shoe …
- `character-reveal` — Ceremonial reveal of the runner's running character based on 9 dimensions computed from real training data. Use when ch…