@urko_garmendia@claudeai Use it for deep research and code base analysis and the difference is very noticeable, deeper insights, finds harder to spot bugs and errors
Beginning July 20, Claude Fable 5 will be included in all Max and Team Premium plans, at 50% of limits.
Pro and Team Standard users will continue to have access to Fable via usage credits, and will receive a one-time $100 credit.
Demand for Fable has been challenging to predict, which is why we rolled it out to subscription plans in stages, extending access several times as we secured additional capacity.
@ericosiu Step 3 is where most of these fall apart in practice. Capturing the memory is the easy part, building in decay so a three year old decision doesn't outweigh last week's is what actually keeps the brain trustworthy.
@gregisenberg Markdown files are the new config is the one people skip. Everyone obsesses over the model and forgets the agent still needs a current, structured place to read its own memory from.
@JoshEdmondson_ Right, markdown is just the easiest format to get there today. The real test is whether the business can explain itself well enough that an agent could pick up the work with no extra context.
@Zephyr_hg We're doing something similar with Workle. Six specialised agents handling content, research, and ops. $500/month for a setup that would've cost me three hires two years ago. The economics are genuinely hard to argue with once you see it running.