FAQ
What is CacheSphere, in one sentence?
CacheSphere helps you compare tradeoffs, transfer concepts, and browse curated records — the same MIT JSON for you and your agents. The 141-language reference is the deepest layer; alongside it sit small hand-curated AI/dev tools (26) and stack records (10), curated models (graduating), preview agent/company/benchmark snapshots, and an experimental, language-oriented Advisor.
What is the difference between a Decision Brief and an agent handoff?
A Decision Brief is the human-readable summary you get from the Advisor: recommended path, viable alternatives, risks, and tradeoffs you can share with your team or paste into a document. A structured handoff is the structured data sent to an AI assistant or agent so it can continue the work with the same constraints, language choices, and stack context. Both are produced from the same Advisor result, but the Brief is for people and the handoff is for machines.
Is it free?
The core site and MIT-licensed data are free to use and self-host. Hosted packaging, support, and paid access are still early-stage; see Pricing & access for the current state.
Is it a course or tutorial site?
No. It is for decision support and orientation: browse catalogue records, compare tradeoffs, and get grounded quickly before you commit to a model, agent, language, tool, or extra subsystem.
Does it replace official docs?
No. CacheSphere is the layer you use before or alongside official docs when you are deciding what to use, what to keep simple, or whether another language is worth the complexity.
Can I use the JSON in my product?
Yes, under the MIT licence. Preserve licence notices and attribute CacheSphere when you quote or closely summarise the data in user-visible output. See llms.txt for agent-oriented guidance.
Will the API work on the public website?
/api/languages.json and the other static JSON catalogues work on the public site. On cachesphere.com,
GET /api/agent/tools and the hosted POST /api/agent/* Worker routes
(select-context-packs, decision-kit, experiment-context) are live. /v1/… routes need the
self-hosted Node server. Details are in Troubleshooting and
For AI tools & agents.
How fresh is the data?
The catalogue improves over time. Treat code and ecosystem notes as grounded starting points, not final authority, especially for fast-moving tools and runtimes. Public curation follows discovery → verification → curation → catalogue.
Who is it for?
Solo developers, small teams, and AI tools that need a better answer than “use the default.” Useful when you are comparing languages, browsing curated catalogues, or justifying a stack change with evidence from the records.
Who runs it?
CacheSphere is built by Idaluna Labs. See About for tiers, methodology, and scope. Contact: [email protected].