Kadhai is in beta. Found a bug, or something confusing? Report it; every report helps.
kadhai
Docs: Settings file (.env)

docs / 16

Settings file (.env)

Keys and models are easiest to manage in the dashboard. Everything else can go in a .env file in the Kadhai folder; copy .env.example to .env and uncomment what you need.

Setting Default Meaning
PORT 4747 Dashboard port (the preview uses the next one)
WORKSPACE_DIR ./workspace Folder used at first start; after that, the folder you open is remembered
GEMINI_API_KEYS, GROQ_API_KEYS, ... Keys, comma-separated, if you prefer a file
PROVIDER_ORDER gemini,groq,cerebras,... Which providers are tried first
MAX_SUBTASKS 6 Most steps in a plan (low flame)
MAX_REVIEW_ROUNDS 1 Review and fix rounds; 0 switches the review off
MAX_OUTPUT_TOKENS 4096 Longest answer per call (low flame)
FILE_CONTEXT_CHARS 24000 File content per prompt (low flame)
EXEC_MODE ask ask, auto or off
EXEC_TIMEOUT_MS 60000 Time limit per command
EXEC_SANDBOX (none) docker runs commands in a container
KADHAI_MOCK 1 starts in demo mode

Numbers you raise here are kept on both flames.

This page comes from the user guide in the repository. Something unclear or wrong? Tell us.

screenshot