Score File

Score an AI session transcript file from disk.

Trigger

Use when the user has an exported session file (from Claude Code, Cursor, Windsurf, or Gemini) and wants to evaluate its orchestration quality.

Workflow

1. •Ask the user for the file path if not provided
2. •Use the score_session_file MCP tool with the file path
3. •Present the full score breakdown with per-metric analysis

Input

A path to a session file. Supported formats:

• •.md — Markdown session exports
• •.txt — Plain text transcripts
• •.jsonl — JSON Lines format

Output

• •Overall score (0-100) with quality tier
• •Per-metric breakdown with bar visualization
• •Signals detected: skills, agents, decisions, errors recovered, compound learnings
• •Complexity classification: simple, moderate, or complex

Constraints

• •File must exist and be readable
• •Empty files or files with no orchestration signals will score low
• •Scoring is offline — no network calls, no data leaves your machine