{
  "name": "reflexion",
  "description": "Collection of commands that force LLM to reflect on previous response and output. Based on papers like Self-Refine and Reflexion. These techniques improve the output of large language models by introducing feedback and refinement loops.",
  "version": "1.0.0",
  "author": {
    "name": "Vlad Goncharov",
    "email": "vlad.goncharov@neolab.finance"
  },
  "commands": [
    "./commands"
  ]
}