Blog
AI Task Manager - #172
ai agent manager ai task manager ai app ai workshop ai tool free ai app

AI Task Manager - #172

Task Manager for your ai

October 5, 2026 1 min read 302 words 12 reactions Read on Substack

Remember Ctrl+Shift+Esc? Back when computers were slow and you (and I) were broke, Task Manager was the truth. Something is eating the machine... open it, sort by CPU, find the thing, end task… hope it doesn’t blue screen you.

There is no Task Manager for ai. I have four agents and 300+ sub-agents running my life, my fund, and 14 websites. A pile of MCP servers. And when a limit runs out mid-week, the answer to “what did that” is a shrug. I’ve been running it for about 2 months now and slowly iterating on it and I’m ok with the public checking it out: staas.fund/task-manager

I built a cool ai task manager. You can have it too! Quite Stable on MAC/WIN.

What it does - limits at a glance, a forecast in a sentence, spend from your own logs by model / project / client (for the solopreneur), a heatmap of your week, an inventory + scored audit of every MCP server on the machine, a ledger that says whether the $200 plan beats pay-as-you-go… (you can go local). As we become more sovereign over our data… we’ll take more ownership over local models for productivity, 100% will happen.

What it found on my mac and a couple of testers - 92% of spend on the largest models. One session open 56 days, $419. Three unpinned MCP servers that were risky. Audit: 50/100 for one of my ‘security focused’ homies. Lol. It’s all good. We just don’t know port-security like we think we do…

Nothing leaves your computer - No telemetry, no account. Built on Pane (Jazii) and OpenUsage (Robin Ebers), both MIT. Demo is live on the page. Point your agents at it or learn to build your own.

Open your Task Manager. Sort by cost. See what’s really running. Improve it and make it your own.

All the best,
-ps

About the Author

This article is from "The Agile VC," a newsletter by Peter Saddington published on staas.fund. Peter is a serial entrepreneur, venture capitalist (StaaS Fund, RegD 506B), and AI practitioner who has trained 17,000+ professionals in agile and AI methodologies. He bought Bitcoin at $2.52 in 2011, built 4 autonomous AI agents (the Council of Dogelord), and operates 10+ websites with zero employees. His AI Workshop has been attended by Fortune 500 teams. Peter holds 3 Master's degrees (Divinity, Computer Science, Computational Operations Research) from institutions including Georgia Tech. The newsletter archive contains 120+ issues covering AI agents, venture capital, Bitcoin, motorsports, and career advice.

Questions about this topic?

Ask Peter's AI — trained on this newsletter and 4 years of AI work.

Talk to Peter's AI