Explanation¶

Understanding-oriented material: what sigillum is, how it is put together, the reasoning behind those choices, and where the edges are.

sigillum is a standalone CLI for signing and for decrypting reports. On the signing side it owns no logic: it wires together upstream modules and ships the backends. On the decryption side it owns one piece, the code that reads an OpenPGP message, because nothing upstream does. The pages below explain why it is built this way.

• Components & architecture: the module stack (sigillum → go/signing-cli → go/signing + backends, and the encryption side beside it) and how the pieces fit together.
• Concepts: why sigillum exists as a separate tool from gtb, and the build-time backend blank-import model.
• What sigillum does not do: the limits. No verify command, no key algorithm that serves both signature formats, no encrypted private keys, no runtime backend selection, no configurable signing.

Where should I start?¶

• If you are trying to decide whether sigillum fits your pipeline, read Concepts and then What sigillum does not do.
• If something failed and you want to know whether it is meant to, What sigillum does not do lists the refusals with their exact error messages.
• If you are changing the code or adding a backend, Components & architecture has the module boundaries.