AI spam classification: the model's opinion as one bounded spam signal, and the llm_prompt Sieve function (AI-1 to AI-28)

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.
This commit is contained in:
2026-09-19 00:41:00 -07:00
parent cba48cf03b
commit 9490fc4677
39 changed files with 2945 additions and 28 deletions
@@ -43,6 +43,7 @@ impl Response<'_> {
GetRequestMethod::VacationResponse(request) => request.resolve_references(self)?,
GetRequestMethod::MaskedEmail(request) => request.resolve_references(self)?,
GetRequestMethod::DeletedAccount(request) => request.resolve_references(self)?,
GetRequestMethod::AiLimits(request) => request.resolve_references(self)?,
GetRequestMethod::Principal(request) => request.resolve_references(self)?,
GetRequestMethod::Quota(request) => request.resolve_references(self)?,
GetRequestMethod::Blob(request) => request.resolve_references(self)?,
@@ -83,6 +84,9 @@ impl Response<'_> {
SetRequestMethod::DeletedAccount(request) => {
request.resolve_references(self, 1, false)?
}
SetRequestMethod::AiLimits(request) => {
request.resolve_references(self, 1, false)?
}
SetRequestMethod::AddressBook(request) => {
request.resolve_references(self, 1, false)?
}