Installation

cargo install sceptre-cli

This is the CLI’s default build: ONNX Runtime is loaded at runtime via ORT_DYLIB_PATH (ort-dynamic) plus model download/caching (download) — runnable out of the box once a compatible libonnxruntime is available.

For a zero-config native binary that bundles a prebuilt ONNX Runtime at build time instead:

cargo install sceptre-cli --features ort-bundled

To also expose the MCP server subcommand, add the mcp feature:

cargo install sceptre-cli --features "ort-bundled,mcp"

The library ships default = [] — no backend and no download support are enabled by default. Enable a backend feature and, typically, download:

[dependencies]
sceptre = { version = "0.1", features = ["ort-bundled", "download"] }

See Feature flags for every flag and when to pick it, and Backends for the ort vs. tract tradeoff.

The MCP surface is behind the mcp cargo feature, exposed as the sceptre mcp subcommand. Install the CLI with mcp enabled (see the CLI tab), or depend on the library with mcp plus a backend and download:

[dependencies]
sceptre = { version = "0.1", features = ["ort-bundled", "download", "mcp"] }

Model download and caching

The CRAFT detector and the gen2 recognizers are first-party ONNX artifacts fetched from Hugging Face (sceptre-ocr/*) on first use. Each download is verified against a pinned SHA-256 before use.

Models are cached in the standard Hugging Face hub cache — HF_HUB_CACHE → HUGGINGFACE_HUB_CACHE → $HF_HOME/hub → ~/.cache/huggingface/hub — so the library, the CLI, and any other Hugging Face tooling on the machine share one cache store. Every run after the first reads the cached models with no network access.