Evidence-based endurance coaching protocol (v11.35). Use when analyzing training data, reviewing sessions, generating pre/post-workout reports, planning work...
**Expanded data source support, tighter data discipline, and multi-platform improvements:** - Clarified data and template file locations for both local and cloud/connector setups, supporting agents (e.g., Claude, ChatGPT, Mistral). - New support for additional JSON sources: intervals.json, ftp_history.json, and routes.json for granular analysis (intervals, DFAs, terrain). - Enforced strict data discipline: every metric cited must come from a JSON read in the current response—no use of prior conversation, memory, or estimation. - Improved agent platform guidance, specifying expected behavior for GitHub connectors, direct file reads, and persistent configuration (SOUL.md, system prompts). - Updated protocol to version 11.35 and revised action hierarchy and fetch precedence for all required files and reports. - Refined first-use setup instructions and directory checks for a smoother onboarding experience.