kobenfang/BigFish

BigFish is an AI skill for angling and outdoor fishing built to follow the cross-platform open Agent Skills standard. It delivers targeted spot analysis, real-time fish activity forecasting, and customized recommendations for fishing methods and bait across multiple types of fishing including lure fishing, still fishing, wild fishing and open sea fishing. It also supports automatic catch logging and generating full reports of completed fishing trips. It works on any agent framework that supports the Agent Skills standard, including DeepSeek Harness, OpenClaw, and Claude Code.

Casual amateur fishing enthusiasts can use BigFish to analyze conditions at their target fishing spot before heading out, combine local weather and barometric pressure data to get tailored suggestions for target species, bait types, and fishing methods, then log all their catches after the trip is complete. Experienced professional anglers can use it to organize their own personal notes and experience from different spots for future reuse and easy sharing with other fishing hobbyists. All you need to do to get a clear result is enter a simple natural language query to the AI agent that has BigFish installed.

BigFish is a fully knowledge-driven skill that only uses four structured YAML knowledge files, so it does not require any extra runtime dependencies or custom scripts to work properly. If you want to add accurate weather data to your fish activity analysis, you can pair it with any existing weather tool that your agent framework already supports. There are two common install methods for the skill: you can clone the git repository to the ~/.agents shared user directory, or copy it to your current project’s .agents directory. It is released under the permissive open source MIT license, so anyone can use it for free.