token-stats

/token-stats - Token Economics Dashboard

Show the token savings achieved by using search-first exploration vs blind file reading.

Instructions

When the user invokes /token-stats, analyze token usage and display savings.

1. Gather Activity Data

REPO_ROOT=$(git rev-parse --show-toplevel 2>/dev/null || echo "$PWD")
ACTIVITY_FILE="$REPO_ROOT/.simba/search/activity.log"

# Count files read this session
if [ -f "$ACTIVITY_FILE" ]; then
    echo "=== Session Activity ==="
    FILES_READ=$(grep -c "READ:" "$ACTIVITY_FILE" 2>/dev/null || echo "0")
    FILES_EDITED=$(grep -c "EDIT:" "$ACTIVITY_FILE" 2>/dev/null || echo "0")
    SEARCHES=$(grep -c "SEARCH:" "$ACTIVITY_FILE" 2>/dev/null || echo "0")
    echo "Files read: $FILES_READ"
    echo "Files edited: $FILES_EDITED"
    echo "Searches performed: $SEARCHES"
else
    echo "No activity log found for this session"
fi

2. Calculate Codebase Stats

REPO_ROOT=$(git rev-parse --show-toplevel 2>/dev/null || echo "$PWD")

# Total files in codebase
TOTAL_FILES=$(rg --files "$REPO_ROOT" 2>/dev/null | wc -l | tr -d ' ')

# Estimate total tokens (rough: ~1.33 tokens per word, ~3 words per line)
TOTAL_LINES=$(rg -c '' "$REPO_ROOT" 2>/dev/null | awk -F: '{s+=$2} END {print s}')
ESTIMATED_TOTAL_TOKENS=$((TOTAL_LINES * 4 / 3))

echo ""
echo "=== Codebase Size ==="
echo "Total indexable files: $TOTAL_FILES"
echo "Total lines: $TOTAL_LINES"
echo "Estimated tokens: $ESTIMATED_TOTAL_TOKENS"

3. Query Memory for Historical Data

uv run python -m simba.search stats

4. Calculate and Display Economics

Based on the gathered data, calculate and present:

Token Economics Model:

Scenario	Calculation	Typical Cost
Blind exploration	Read 20+ files to find relevant code	~50,000 tokens
With turbo-search	Search (50 tokens) + Read 3-5 targeted files	~5,000 tokens
Savings		~90%

Present this table to the user: