b11fcb8a7ff51bf2218a76d7ba98ec609e05726a
3
Commits
| Author | SHA1 | Message | Date | |
|---|---|---|---|---|
  GustavoandClaude Sonnet 5
|
124516e64a |
fix(v2): fix "Validar e executar" bug (real root cause), expand generic-phrase blacklist, add FORCE_SEND
Root cause found for why the real Slack message showed "→ Validar e
executar" for nearly every ACTIONABLE_NOW ticket: logic/message.js's
ACTION_LABELS map had a hardcoded entry for ACTIONABLE_NOW that always
won over the AI's actual next_action field in shortActionLabel(). The
AI's real, specific, already-validated-as-non-generic next_action (e.g.
"Realizar inventário das máquinas") was being silently discarded for
display purposes on every single message. Fixed by removing the
ACTIONABLE_NOW entry -- it now falls through to the existing
next_action-based fallback (already there, just unreachable for that
class before). UNASSIGNED_ACTIONABLE intentionally keeps its fixed
"Atribuir técnico" label -- that IS the correct display for that class.
logic/classification.js: expanded GENERIC_NEXT_ACTION_PHRASES with the
exact phrase that leaked through ("validar e executar") plus the
mission's additional list (validar chamado, prosseguir atendimento,
executar chamado, tratar chamado, analisar chamado). Bumped
AI_CLASSIFIER_VERSION 3->4 since this changes what next_action values
validate as non-generic -- forces reclassification of any cached
candidate that only passed under the older, narrower blacklist.
n8n/nodes/validate-ai-response.node.js: generic-next-action rejections
now get their own distinct fallback_reason ('generic_or_unsupported_
next_action') instead of being lumped into 'invalid_json', matching the
same pattern already used for the technician-vendor conflict.
logic/slack.js + logic/config.js: added FORCE_SEND (default false).
Manual-triggered runs now never send for real even with SEND_ENABLED=
true in production, unless FORCE_SEND is also explicitly set -- protects
against someone clicking "Execute Workflow" in production and
accidentally re-sending today's digest.
Added format-safety regression tests (no literal backslash-n, URL never
bleeds into the next item's id, no markdown-wrapped URLs) -- verified
these were never actually triggered in a real send (the reported
corruption was n8n's raw JSON data-inspector view, not a delivered
Slack message; both manual runs in question had shouldSend:false).
169/169 tests passing.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
|
||
|
GustavoandClaude Sonnet 5
|
482b7ef48e |
fix(v2): redefine AI actionability semantics + version the classifier (AI_CLASSIFIER_VERSION)
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> |
||
|
GustavoandClaude Sonnet 5
|
a99a25cb1a |
feat(v2): complete n8n workflow template (36 nodes) covering full V2 pipeline + shadow branch
- 14 new Code node drivers implementing the pipeline stages: normalize
candidates, build context SQL, merge context + hash, decide AI need,
prepare AI prompt, validate AI response, AI-disabled fallback, reuse
cached classification, compute score, select top 10, prepare memory/
envio records, build message/digest, compare legacy x V2, debug panel.
- Consolidated the two context queries (followups/tasks) into a single
UNION ALL query tagged by tipo, and added a "Merge - Unir Shadow"
synchronization node before comparing legacy x V2: n8n does not
guarantee relative execution order between sibling branches that only
share an early common ancestor, so cross-branch data was only safe to
read by name once a real Merge node forced both to complete first.
- Reordered score calculation before the memory upsert (score is needed
in the persisted row) and turned both Data Table writes into
side-effect branches that fan out from the main path instead of
gating it, so the Data Table node's own output shape never needs to
carry the full pipeline state forward.
- Fixed a duplicate-identifier bug in the build script: score.js's
require('./cooldown') survived inlining and collided with
cooldown.js's own computeCooldown function in the same Code node
scope. build-workflow.js now strips inter-module requires when
inlining, and a new test syntax-checks every generated jsCode string
plus asserts no leftover require() calls.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
|