Launching TokenDam tomorrow — a linter for your LLM token spend

Hey everyone — I'm launching TokenDam here tomorrow and wanted to share the story a day early.

I kept staring at agent token bills that felt too high but couldn't see why. Turns out the waste is boringly consistent: the same 2-3k-token system prompt resent uncached on every call, a whole scraped page dumped into context, 20+ MCP tools the model never touches, history that never gets compacted.

TokenDam reads a trace (an OpenAI or Anthropic request payload) and points at each leak with a dollar figure and a concrete fix. It runs 100% in your browser — your prompts never leave the tab — and there's a free CLI too (npx tokendam). The paid Cloud tier is a GitHub App that audits every PR for token-cost regressions.

If you run agents in production, I'd genuinely love your take: what's the biggest surprise you've found in your own token bill? And what would make a tool like this part of your workflow vs. a one-time audit?

Try it: https://tokendam.dev — would love your feedback before tomorrow.