[Release] Storytime - A continuity system for claude code
I recently released my claude code plugin, Storytime(github). It's been underwhelming to say the least, but its more of a relief than a joy to put it out.
I harbor a lot of ideas at any given point, and they turn into "what-if" more than "this will be amazing". I think thats better TBH.
Storytime was my personal playground to explore a giant pile of "what-if" gaps in the Claude Code (CC) functionality. Turns out the Anthropic team caught up and surpassed my plugin in the time it took to sound it out.
There's still a ton of valuable ideas in there to explore, but only a few are worth using day-to-day. The spec system is overbuilt. The roles/personas are weak. The icebreaker/breakout/converge system is hard to discern if its better or worse.
I think the biggest question for me right now is how CC functionality is better or worse than any given plugin, on the longtail. We have tons of initiatives that say "from one prompt" or start from zero, but what about brownfield 20 or 100 prompts down the chain? How does that chain wind back up and maintain its usefulness?
The leak earlier this year of the CC code is telling in how they have a sophisticated and bespoke internal set of systems that are focused around exactly these problems. As we saw with Fable recently, they move more to removing the human, and ignoring important reflection points to add nuance and clarity. Wow, the first time with Fable was an absolute stallion of a rodeo (but thats another post i dont care to write, Simon Willison wrote it better).
So where does that leave Storytime? I'm going to use this opportunity to step back and cut it down into discrete modules. Its probably stronger in non-CC environments, if at all. In the meantime, I want to loop me more into the decision-making process, and so will continue to build more targeted and effective skills that keep me building what I intend, not what the model decides to vibe at me.