Assess Decide Do in Claude, but with Skills
How using Claude skills reduced my ADD framework's integration overhead by 94%. From monolithic megaprompt to dynamic, modular architecture. Open source.
Building addTaskManager, the best iOS productivity app for ADHD minds
How using Claude skills reduced my ADD framework's integration overhead by 94%. From monolithic megaprompt to dynamic, modular architecture. Open source.
Learn to spot sophisticated LinkedIn recruitment scams that use compromised accounts and malicious repos. Use AI tools like Claude Code to scan code before running.
Reveals AI that calls itself "appropriately modest" is slightly chilling. Exploring the inflection point where machine-validated truth starts feeling indistinguishable from reality.
Claude Cowork replaces traditional apps with natural language. See real examples of file organization and image creation done in seconds instead of minutes.
What is bio content and why 2026 might be its breakout year. I believe 2026 will be the year of "bio content", a.k.a. human generated content, [...]
AI eliminates cognitive friction that built our skills. Explore what happens when delayed gratification disappears and instant results become the norm.
Andrej Karpathy's LLM Council already follows the Assess-Decide-Do pattern. I made it explicit with a PR that adds cognitive scaffolding. Here's how it works.
The crypto mantra "don't trust, verify" now applies to all digital life. Practical filters for surviving deepfakes, bots, and AI-generated content.
2025 year in review: cleaning up old projects, adding new tools and getting ready for a challenging 2026.
AI creates plausible realities as addictive as drugs. Someone who lived under communism and capitalism sees a new ideological split emerging from this shift.
AI is trained on past patterns and can't recognize disruptive excellence. Like critics who booed Fight Club at Cannes, it will always miss what truly matters.
We watched The Matrix, understood the warning about sentient programs, and then built AI agents anyway. Four scenarios for how this could go wrong.
Open-source AI tools / Claude skills for content creators: link analysis, batch SEO updates, and smart CTA injection. Built during a real migration, now open source.
AI can amplify your work, but only if you're already doing things worth amplifying. Learn where AI actually helps, and what is the hidden trap of [...]
Integrating a human cognitive framework in Claude makes the LLM partially "understand" humans. World models may be better suited for this.
Re-evaluating the Assess-Decide-Do framework in the age of AI. Not all life areas are suitable, but in most of them, it fits perfectly.
15 years ago I built the ADD framework. Now I taught it to Claude itself. The result: unexpected empathy. Full guide + mega prompt.
In 2019 I predicted 10 AI disruptions. Some came true, others not, and I missed ChatGPT. Here's what getting it wrong taught me about change.
AGI won't be a single massive system but many specialized agents working together: swarm intelligence, energy efficiency, and distributed AI.
Large language models (ChatGPT included) and astrology are using the same approach. One uses compute, the other uses collective human memory.