All numbers pulled live from fleet.json (refreshed every 5 min by a cron on the Mac Studio running the fleet). Source breakdown:
- Tasks done — counted from per-platform output logs (tweets posted, videos uploaded, PRs merged, emails drafted, etc.). Verified.
- Tokens processed — from
~/.openclaw/cron/runs/*.jsonl where each run records its model + input/output token count. Verified for ~12k cron runs. - LLM cost ($0) — the cron pipeline runs almost entirely on local Ollama models (qwen2.5, qwen3, gemma3) on the Mac Studio. No API marginal cost.
- Anthropic API spend — Rico (the code-writing agent) uses Claude via the Anthropic API. That spend lives in console.anthropic.com — last verified at ~$50 across all Anthropic usage org-wide (incl. dev sessions building the fleet). Not yet auto-pulled into this dashboard.
- Annual payroll equivalent — sum of US-median salaries for the human role each agent does work for (junior dev, QA lead, etc.). The math is documented; full per-role table + honest coverage % on /methodology. Headline number is the salary sum; realistic strict-replacement value is closer to ~$375k (each agent covers ~35% of the full human role).