asksurf-ai/surf-skills

Surf-skills targets the 'agentic workflow' trend by wrapping various crypto data providers into a format easily consumable by LLMs (likely as tools/functions). With 24 stars and 1 fork in 44 days, adoption is currently low. The project's primary value is the 'convenience' of not having to map 83+ endpoints manually. However, this is a thin layer with low defensibility. The moat is purely the maintenance of these API mappings. Competitors include established data aggregators like Birdeye, CoinGecko, or Dune, who are increasingly launching their own official 'Agent SDKs' or MCP (Model Context Protocol) servers. If a major provider like Coinbase or Alchemy releases an official LLM-optimized toolkit, third-party wrappers like this will face immediate obsolescence. The displacement horizon is short (6 months) because the technical barrier to replicating this via an LLM generating its own API wrappers is rapidly decreasing.