stella init

Analyze the current workspace, infer its domain taxonomy, and build the code-graph index that powers memories, reflections, and structured exploration.

Synopsis

stella init [global flags]

What it does

stella init does the following:

1. Infers the domain taxonomy and writes it to <workspace>/.stella/domains.toml. This taxonomy is the tagging vocabulary for memories, reflections, and every code-graph node and edge. It is also persisted into the context plane at <workspace>/.stella/context.db.
2. Builds the code-graph index at <workspace>/.stella/codegraph.db — the database stella graph queries.
3. Syncs extensions — adopts commands, skills, and agents from other agents' .claude/ and .agents/ directories (workspace and user scope) as symlinks under .stella/.

It works offline using a heuristic fallback, so it never hard-fails on configuration — you can initialize a workspace even without a resolvable provider credential.

Flags

stella init takes the global flags. Because it can run offline, model selection is optional:

The global --budget flag is accepted but not honored by init — the analysis makes at most a couple of token-capped inference calls and does not track a budget.

Generated files

Examples

Initialize the current workspace:

stella init

Initialize using a specific model for taxonomy inference:

stella --model openai/gpt-5.5 init