Skip to content

Latest commit

 

History

History
60 lines (49 loc) · 2.8 KB

install.md

File metadata and controls

60 lines (49 loc) · 2.8 KB

Installation

npm install -g autoillustrate    # or: npx autoillustrate ...
autoillustrate doctor            # checks every prerequisite and prints fix-it hints

Security note — AutoIllustrate launches coding agents that can run arbitrary commands and inherit your environment, including your API keys. The source document you pass in steers those agents, so treat it like code: only run it on papers and files you trust, and preferably inside a container or VM rather than directly on your workstation. Run logs under <workdir>/logs/ can echo document contents, so review them before sharing a run directory.

Prerequisites

  • draw.io desktop CLI on PATH as drawio. Install the draw.io desktop app; the CLI binary is bundled with it.
  • Node ≥ 22.
  • Python ≥ 3.10 — no manual setup is needed. On first run, the CLI creates a cached venv under ~/.cache/autoillustrate/. Set VENV_PY=/path/to/python to use your own environment instead.
  • Headless CLI (@roberttlange/headless) — all agent calls go through it. It is installed with AutoIllustrate. The default agent is Claude Code, so claude must also be on PATH unless you select a different --agent.
  • OPENAI_API_KEY — required by default. Phase 3 and Phase 6 vision audits use gpt-6-astra, and the Phase 5 restyle uses gpt-image-2.5-sunburst unless a different --enhance-model is selected.
  • GEMINI_API_KEY — required only for a Gemini backend: --enhance-model nanobanana|nanobanana-pro|gemini-* or --audit-vlm gemini-3.1-pro.

Either export the keys in your shell or place them in a .env file in the directory you run autoillustrate from. autoillustrate credentials reports which keys and agent logins the pipeline can see.

draw.io on a headless Linux box

On a Linux machine with no display, run the install_drawio.sh script that ships inside the package (autoillustrate doctor prints its full path). It installs a pinned drawio via Nix and wraps it in xvfb-run so drawio -x works without an X server.

Chrome for the export fallback

drawio's own PNG export occasionally wedges on formula-heavy figures. When a Chrome or Chromium binary is available (auto-detected, or set AUTOILLUSTRATE_CHROME), the pipeline falls back to rasterizing an SVG export with it. Without Chrome the fallback is unavailable; see Export reliability.

Other agents

The --agent flag switches the coding agent that runs the LM sub-agents: claude (default), codex, cursor, gemini, opencode, or pi. The chosen agent's CLI must be on PATH and logged in; autoillustrate doctor --agent <name> checks it. See the CLI reference for per-agent model and effort defaults.