CloudBuddy Blog
Practical writing on workflow automation, AI systems, and how businesses eliminate manual work.
Anthropic released Fable 5 today, a Mythos-class model anyone with an API key can use. I ran it head-to-head against Opus 4.8 and Sonnet 4.6 on launch day, from quick puzzles to a 150,000-line codebase audit. The frontier difference shows up exactly where you'd expect, and not where you wouldn't.
Read post
The easiest way to understand useful AI at work is to look at the queue: the inbox, ticket list, approval pile, lead backlog, or report stack where work waits for the next step.
Read post
AI made software cheap to produce, which means the software itself is no longer the moat. Four advantages still compound: speed of the full loop, anything anchored to the physical world, data that improves with use, and owning one narrow problem end to end.
Read post
Most AI projects fail at the first decision: which workflow to build for. AI Workflow Ranking is a repeatable way to score every workflow on readiness and value, then pick the first build that actually pays off.
Read post
The best agents are not magic chat windows. They are narrow workflow systems with clear purpose, typed tools, evals, human review, and staged autonomy.
Read post
A repeatable framework for auditing a SaaS codebase at scale. A set of audit tracks you select and adapt to your system, an invariants loop that stops regressions, and a verification cycle that makes each audit cheaper than the last. One recent application surfaced hundreds of findings and promoted 36 invariants to code-level guardrails.
Read post
Anthropic released Mythos Preview to a closed group of organizations. The capability leap is real, but the access model may matter just as much. As AI shifts from an equalizer to a gated advantage, the teams that win will be the ones that can turn that capability into working systems.
Read post
Claude Code CLI accidentally exposed part of its codebase. I pulled the package and looked under the hood. The direction is clear: AI agents are becoming systems.
Read post
What it actually looks like to work with remote AI agents. What worked, what didn't, and how Claude Dispatch compares to our custom AI tools.
Read post