strava-api
Strava API
Direct integration with Strava API v3 for fitness data management.
Security Model
IMPORTANT: All secrets are stored in system keychain (macOS Keychain / Linux Secret Service).
- AI agents CANNOT read secrets from keychain
- Only the Python scripts can access credentials at runtime
- Non-secret metadata (athlete name, token expiry) stored in
~/.strava/metadata.json
This design ensures your API credentials are never exposed to AI agents.
First-Time Setup Workflow
When user requests Strava data for the first time:
Step 1: Check if configured
python3 scripts/strava_api.py athlete
If error "Strava not configured" → proceed to Step 2. If success → skip to "Using the API" section.
Step 2: Create Strava Application
Guide user to create API application:
- Open https://www.strava.com/settings/api
- Fill application form:
- Application Name: Any name (e.g., "My Fitness App")
- Category: Choose any
- Website:
http://localhost - Authorization Callback Domain:
localhost
- Note Client ID and Client Secret (will be entered in setup)
Step 3: Run Interactive Setup
python3 scripts/setup_oauth.py
Script will:
- Ask for Client ID and Client Secret (secret input hidden)
- Open browser for Strava authorization
- Guide user to copy redirect URL
- Exchange code for tokens automatically
- Store secrets in system keychain (secure)
- Store metadata in
~/.strava/metadata.json - Verify connection
Step 4: Verify Setup
python3 scripts/strava_api.py athlete
Should display athlete profile JSON.
Using the API
CLI Commands
# Get athlete profile
python3 scripts/strava_api.py athlete
# Get athlete stats
python3 scripts/strava_api.py stats
# List recent activities
python3 scripts/strava_api.py activities --limit 20
# Get specific activity
python3 scripts/strava_api.py activity 12345678
# Explore segments in area
python3 scripts/strava_api.py segments "37.7,-122.5,37.8,-122.4" --type running
# Raw API request
python3 scripts/strava_api.py raw GET /athlete/zones
python3 scripts/strava_api.py raw POST /activities --data '{"name":"Test","sport_type":"Run","start_date_local":"2024-01-15T10:00:00Z","elapsed_time":1800}'
Python Import
from scripts.strava_api import StravaClient
client = StravaClient() # Auto-loads from keychain, refreshes token if needed
# Get athlete
athlete = client.get_athlete()
# List activities
activities = client.list_activities(per_page=50)
# Create activity
new_activity = client.create_activity(
name="Morning Run",
sport_type="Run",
start_date_local="2024-01-15T07:30:00Z",
elapsed_time=1800,
distance=5000
)
Token Management
Tokens refresh automatically. Check status:
# Check token status
python3 scripts/refresh_token.py --status
# Force refresh
python3 scripts/refresh_token.py --force
Decision Tree
User Request
├── First time / "setup strava" → Run setup_oauth.py
├── "Get my profile/stats" → strava_api.py athlete/stats
├── "List my activities" → strava_api.py activities
├── "Get activity details" → strava_api.py activity {id}
├── "Create/log activity" → client.create_activity()
├── "Update activity" → client.update_activity()
├── "Get training data/streams" → client.get_activity_streams()
├── "Find segments nearby" → strava_api.py segments {bounds}
├── "Export route" → raw GET /routes/{id}/export_gpx
└── "Token expired" → Auto-handled by client
Quick Reference
| Item | Value |
|---|---|
| Secrets storage | System Keychain (secure) |
| Metadata location | ~/.strava/metadata.json |
| Base URL | https://www.strava.com/api/v3 |
| Rate limits | 100 req/15min, 1000 req/day |
| Token lifetime | 6 hours (auto-refresh) |
References:
- Full API docs: references/api_reference.md
- Manual setup: references/setup_guide.md
Sport Types
- Running:
Run,TrailRun,VirtualRun - Cycling:
Ride,MountainBikeRide,GravelRide,EBikeRide - Other:
Swim,Walk,Hike,Workout,WeightTraining,Yoga
Error Handling
| Status | Meaning | Action |
|---|---|---|
| 401 | Token expired | Auto-handled by client |
| 403 | Insufficient scope | Re-run setup_oauth.py |
| 404 | Resource not found | Check ID |
| 429 | Rate limit | Wait 15 min |
More from kdoronin/claude_code_skills
text-to-sql
|
15skill-orchestrator
This skill orchestrates programming tasks by analyzing available Claude Code skills and creating execution plans. It should be used when working on any coding task that could benefit from multiple specialized skills. The skill supports two execution modes selected by user - manual (Claude executes with explicit skill references) or delegated (tasks sent to sub-agents with skills).
6plugin-creator
Comprehensive plugin development skill for Claude Code. Analyzes business requirements and creates complete plugins including MCP servers, skills, slash commands, or combinations thereof. Supports TypeScript, Python, and any other language. Provides architecture guidance, templates, and end-to-end implementation.
4project-agent-bootstrap
Bootstrap new or existing local projects for AI-assisted development by creating an `AGENTS.md` for Codex or a `CLAUDE.md` for Claude Code from the bundled rules template, wiring project-local MCP servers for Playwright and shadcn/ui, and initializing Git with a first checkpoint commit when needed. Use whenever the user asks to start a project with agent instructions, scaffold a repo for Codex or Claude Code, add local MCP servers to a project, or standardize an AI-ready project workspace before implementation begins.
1