runstr-fitness_skill

Overview

This skill gives an AI agent read access to encrypted RUNSTR fitness backups stored on the Nostr network so the agent can use your real workout, habits, journal, mood, and step data for coaching and insights. It uses your Nostr private key (nsec) only to decrypt the backup in-session and does not retain the key. The skill enables personalized fitness summaries, trend analysis, habit accountability, and coaching actions based on your actual data.

How this skill works

The agent decodes the provided nsec to derive your public key, queries designated Nostr relays for a RUNSTR encrypted backup (kind 30078), and decrypts and decompresses the payload. It also optionally fetches legacy public workout events (kind 1301) and merges them with backup data. The skill then computes summaries, trends, habit status, and coaching suggestions and stores a compact health summary in memory for follow-up conversations.

When to use it

• When a user asks for a summary of recent workouts, distances, or calories.
• When a user wants coaching or training recommendations based on real history.
• To check habit streaks, journaling trends, mood correlations, or step averages.
• When validating progress toward a goal like weekly distance or frequency.
• When the user asks to sync or refresh fitness data from RUNSTR backups.

Best practices

• Ask the user to provide their RUNSTR nsec only for the current session; never store it.
• Warn users if the backup export date is stale and suggest re-backup in the app.
• Always query the four default relays for best backup coverage.
• Merge and deduplicate backup and legacy public workouts by start time or ID.
• Store a small structured summary in memory rather than re-querying on every turn.

Example use cases

• Generate a 30-day workout summary with activity breakdowns and personal bests.
• Recommend a three-week run plan using recent pace and weekly volume.
• Notify the user that their smoking-abstinence streak dropped and suggest check-ins.
• Correlate mood entries with workout frequency to highlight patterns.
• Compute weekly step averages and suggest small daily goals to bridge gaps.