Stemmacompiler
Install Stemma

Docs / Getting started

Quick start

Go from an existing agent configuration to one canonical project and four native targets. Nothing below reads source code, contacts a service or asks a model to rewrite your words.

  1. 1

    Scan your repository

    List the supported configuration Stemma can see. Scan is read-only.

    shell
    stemma scan
  2. 2

    Import existing guidance

    Choose one detected provider as the source and name every target you want to generate. This creates .stemma/ and records provenance for each imported entity.

    shell
    stemma import --from github-copilot --targets github-copilot,claude,codex,kiro
  3. 3

    Review, then apply

    Plan is read-only. It shows destinations and projection outcomes before apply writes the provider-native files transactionally.

    shell
    stemma plan --allstemma apply --all --yes

Your canonical project

Edit once, compile everywhere.

The canonical source is ordinary Markdown plus bounded front matter. Provider files are generated delivery artifacts; make future edits under .stemma/ and run stemma apply --all --yes again.

Learn the canonical model →

.stemma/ 
.stemma/├── project.json      # name, targets, token budgets├── context/          # architecture.md, testing.md, …├── rules/            # api-validation.md, …├── procedures/├── skills/├── agents/├── decisions/├── profiles/         # <target>.json — delivery, never truth├── provenance.json   # where each entity came from└── manifest.json     # files generated, or verified at import

Starting from nothing?

Run stemma init, author the first canonical entity, then use the same plan and apply loop.