skill-bartender: Minimal Skill Allocation — Task-Skill Matching for DeepSeek Harness (Meta-skill)

# Introduction DSH's philosophy is that everything is a plugin. After loading many skills, the agent faces a constantly growing directory. While it can see the name and description of each skill, the common behavior is "over-dumping": loading too many skills, loading the wrong skills, or missing the workflow skill that was originally intended to string the whole task together. This presents an asymmetric cost: once a wrong skill is loaded, the skill content remains in the conversation history; whereas missing a load only results in the loss of a single tool call. Since missing is cheaper than being wrong, a reasonable strategy is to err on the side of loading less. skill-bartender is implemented according to this logic—it does not provide specific capabilities, but is only responsible for deciding "which skill this task should load, or whether to load it at all."