Configuration¶

greybeard stores its config at ~/.greybeard/config.yaml.

Interactive setup¶

greybeard init

View current config¶

greybeard config show

Set a value¶

greybeard config set <key> <value>

Available keys¶

Key	Description	Example
`llm.backend`	LLM backend to use	`openai`, `anthropic`, `ollama`, `lmstudio`
`llm.model`	Model name (overrides backend default)	`gpt-4o-mini`, `llama3.2`
`llm.base_url`	Custom API base URL	`http://localhost:11434/v1`
`llm.api_key_env`	Env var to read API key from	`MY_OPENAI_KEY`
`default_pack`	Default content pack	`staff-core`, `oncall-future-you`
`default_mode`	Default review mode	`review`, `mentor`, `self-check`, `coach`

Examples¶

# Switch to Ollama (local, free)
greybeard config set llm.backend ollama
greybeard config set llm.model llama3.2

# Use a cheaper OpenAI model
greybeard config set llm.backend openai
greybeard config set llm.model gpt-4o-mini

# Default to mentor mode
greybeard config set default_mode mentor

# Default to on-call pack
greybeard config set default_pack oncall-future-you

Config file reference¶

# ~/.greybeard/config.yaml

default_pack: staff-core # default content pack
default_mode: review # default review mode

llm:
  backend: openai # openai | anthropic | ollama | lmstudio
  model: gpt-4o # leave empty to use the backend's default
  base_url: "" # leave empty to use the backend's default
  api_key_env: OPENAI_API_KEY # env var name (not the key itself)

pack_sources: [] # remote sources to keep synced (future feature)

API keys¶

greybeard reads API keys from environment variables — it never stores them in the config file.

# Set in your shell profile (~/.bashrc, ~/.zshrc, etc.)
export OPENAI_API_KEY=sk-...
export ANTHROPIC_API_KEY=sk-ant-...

Or use a .env file in the directory where you run greybeard:

# .env
OPENAI_API_KEY=sk-...

Local backends

Ollama and LM Studio don't require an API key — just a running server.