jcoffey-dev
9490fc4677
The classifier sends only the subject and text, between unforgeable markers after the operator's prompt, to an OpenAI-compatible endpoint the operator configured; nothing is preset. Its answer maps to an LLM_ tag whose score is clamped (+5.0, -1.0 by default) and can never discard or reject on its own; X-Spam-LLM is sanitized, encoded and folded, and a planted one is removed. Failures, timeouts past the ceiling, a full slot or a paused model leave mail flowing untagged. llm_prompt answers trusted scripts, and accounts holding interactAi within an hourly limit. Redirects aren't followed and no content or secret is logged. The limits live in inbuxa:AiLimits. Acceptance tests 1 and 3 to 21; test 2 as the re-enabled shared llm case, whose setup no longer waits on a rules file from a developer's own path; test 22 written as the ignored ai_compat.
26 lines
839 B
Rust
26 lines
839 B
Rust
/*
|
|
* SPDX-FileCopyrightText: 2026 Coffey Labs
|
|
*
|
|
* SPDX-License-Identifier: AGPL-3.0-only
|
|
*/
|
|
|
|
//! INBUXA's rebuilt features.
|
|
//!
|
|
//! Upstream ships these only in its Enterprise Edition. INBUXA rebuilds each
|
|
//! one clean-room from a written spec under `docs/spec/features/`, one module
|
|
//! per feature, and ships it to everybody (docs/spec/SPEC.md §2.3, §3).
|
|
//!
|
|
//! Upstream files change only by small hooks that call in here, each marked
|
|
//! with an `inbuxa:` comment naming the requirement it serves. That keeps
|
|
//! every upstream merge's conflicts few and predictable.
|
|
//!
|
|
//! This crate sits below `common`, so hooks anywhere in the server can call
|
|
//! it. It works on registry objects and the store directly, never on
|
|
//! `common::Server`.
|
|
|
|
pub mod ai;
|
|
pub mod branding;
|
|
pub mod masked_email;
|
|
pub mod tenancy;
|
|
pub mod undelete;
|