Production-readiness cycle. Classifier was excessively conservative (real Run A: ACTIONABLE_NOW/UNASSIGNED_ACTIONABLE both 0) and made concrete semantic mistakes on real tickets: - #34419: technician "Rafael Brito" (internal) classified as WAITING_VENDOR waiting_for -- an internal technician is not a vendor. - #37389: ticket had concrete open tasks ("Realizar um inventário das máquinas", "Alinhar o dia...") but was classified INSUFFICIENT_CONTEXT. - Generic next_action ("Revisar chamado") repeated across unrelated tickets, proving nothing about real actionability. Rewrote the system prompt (n8n/nodes/prepare-ai-prompt.node.js) with: practical actionability definition ("suporte consegue fazer algo util agora", explicit inclusion list); contatar-vs-aguardar distinction with examples; WAITING_* requires positive evidence (action already taken AND external dependency still open); technician-vs-vendor distinction citing the TECNICO context field; STATUS Planejado alone != blocked/scheduled; SCHEDULED requires an explicit date/window; mandatory chronological interpretation (most recent apontamento wins over a superseded old blocker); next_action must be concrete, with explicit accepted/rejected examples; INSUFFICIENT_CONTEXT bar raised (ask "is there a diagnostic step available even without full info" first). n8n/nodes/prepare-ai-prompt.node.test.js: 10 new tests asserting each rule's text is still present in the prompt (deterministic proxy -- behavioral proof comes from the real DEV run, inspected manually). logic/classification.js: added AI_CLASSIFIER_VERSION=3 (stamped on every valid classification); added a deterministic safety net (not just a prompt hope) -- isWaitingForInternalTechnician() rejects WAITING_VENDOR when waiting_for matches the ticket's own TECNICO field, routing to a new fallback_reason 'ai_semantic_conflict' instead of accepting a known-wrong semantic classification; added next_action genericness rejection (isGenericNextAction) for ACTIONABLE_NOW/UNASSIGNED_ACTIONABLE only. n8n/nodes/validate-ai-response.node.js: now passes the ticket into validateAiClassification() so the technician guard has the facts it needs. logic/cache.js: added buildCacheKey(contextHash, classifierVersion) -> "<hash>:v<version>", stored in the same last_context_hash column (no schema migration) -- bumping AI_CLASSIFIER_VERSION now forces a cache miss on every ticket without manually clearing the Data Table. n8n/nodes/decide-needs-ai.node.js / prepare-memoria-analise-record.node.js updated to compute/persist the composite key. logic/config.js: AI_MAX_CANDIDATES default raised 30->60 (now equals CANDIDATE_POOL_SIZE) -- no candidate should go unanalyzed by an artificial cap now that Run B already proved the cache makes repeat runs cheap; AI_FALLBACK stays reserved for real provider failures. Also added SEND_ENABLED (default false) and ENVIRONMENT config fields for the real Slack send gate (used by logic/slack.js, next commit). Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
70 lines
3.0 KiB
JavaScript
70 lines
3.0 KiB
JavaScript
const { test } = require('node:test');
|
|
const assert = require('node:assert/strict');
|
|
const fs = require('node:fs');
|
|
const path = require('node:path');
|
|
|
|
// O prompt em si nao pode ser testado deterministicamente sem chamar a IA de verdade -- estes
|
|
// testes garantem que as REGRAS exigidas pela missao "finalizacao para producao" continuam
|
|
// escritas no prompt (nao foram removidas/reescritas por acidente numa edicao futura). A prova
|
|
// comportamental real fica nas execucoes reais contra o GLPI (Run com AI_CLASSIFIER_VERSION
|
|
// nova), inspecionadas manualmente no relatorio.
|
|
function readPrompt() {
|
|
return fs.readFileSync(path.join(__dirname, 'prepare-ai-prompt.node.js'), 'utf8');
|
|
}
|
|
|
|
test('prompt define acionabilidade de forma pratica ("suporte consegue fazer algo util agora")', () => {
|
|
const src = readPrompt();
|
|
assert.ok(/suporte consegue fazer algo util agora/i.test(src));
|
|
});
|
|
|
|
test('prompt distingue contatar (ainda nao feito) de aguardar (ja feito) -- teste obrigatorio #2/#3', () => {
|
|
const src = readPrompt();
|
|
assert.ok(/AINDA NAO foi contatado/.test(src));
|
|
assert.ok(/ja foi contatado/i.test(src) || /Ja ter contatado/.test(src));
|
|
});
|
|
|
|
test('prompt exige evidencia dupla (acao ja executada + dependencia externa aberta) pra WAITING_*', () => {
|
|
const src = readPrompt();
|
|
assert.ok(/evidencia POSITIVA/.test(src));
|
|
assert.ok(/dependencia externa ainda aberta/.test(src));
|
|
});
|
|
|
|
test('prompt instrui que tecnico interno (campo TECNICO) nunca vira WAITING_VENDOR -- teste obrigatorio #1', () => {
|
|
const src = readPrompt();
|
|
assert.ok(/NUNCA classifique como WAITING_VENDOR/.test(src));
|
|
assert.ok(/tecnicos internos da equipe de suporte/.test(src));
|
|
});
|
|
|
|
test('prompt instrui que STATUS Planejado sozinho nao significa bloqueio -- teste obrigatorio #6', () => {
|
|
const src = readPrompt();
|
|
assert.ok(/Planejado sozinho NAO significa/.test(src));
|
|
});
|
|
|
|
test('prompt exige data/horario/janela explicita para SCHEDULED', () => {
|
|
const src = readPrompt();
|
|
assert.ok(/SCHEDULED apenas quando existir data, horario, janela/.test(src));
|
|
});
|
|
|
|
test('prompt exige interpretacao cronologica (apontamento mais recente prevalece) -- teste obrigatorio #8', () => {
|
|
const src = readPrompt();
|
|
assert.ok(/ordem temporal/.test(src));
|
|
assert.ok(/mais recente relevante prevalece/.test(src));
|
|
});
|
|
|
|
test('prompt rejeita next_action generico com exemplos explicitos', () => {
|
|
const src = readPrompt();
|
|
for (const frase of ['Revisar chamado', 'Dar andamento', 'Analisar situacao', 'Verificar chamado', 'Acompanhar']) {
|
|
assert.ok(src.includes(frase), `prompt deveria citar "${frase}" como exemplo generico rejeitado`);
|
|
}
|
|
});
|
|
|
|
test('prompt pede para considerar acao de diagnostico antes de usar INSUFFICIENT_CONTEXT -- teste obrigatorio #7', () => {
|
|
const src = readPrompt();
|
|
assert.ok(/Falta de diagnostico nao significa falta de acao/.test(src));
|
|
});
|
|
|
|
test('prompt continua exigindo resposta em JSON estrito com o dominio de classes correto', () => {
|
|
const src = readPrompt();
|
|
assert.ok(src.includes('VALID_CLASSIFICATIONS.join'));
|
|
});
|