MVP M8: background detection — assumptions, risks, inconsistencies
The editor now runs Socrates' three Phase-0-validated detection prompts against the live model, persists the findings, and surfaces them in a new FindingsPanel beside the IssuesPanel. Click any finding to focus its linked element across rail + diagram. Re-detect after model edits to refresh against the new state. apps/web/lib/llm/prompts/socrates - detect-assumptions.md / detect-risks.md / detect-inconsistencies.md promoted verbatim from phase-0 (Phase 0 corpus validated them 10/10). apps/web/lib/llm/detect.ts - Three sequential detection passes (parallel was OOM-prone on 4B local models — Phase 0 lesson). Each pass uses the Phase 0 JSON schema with jsonObjectMode fallback + chatJSON repair-retry. Fail-soft per pass: one busted pass returns [] rather than blowing up the whole detect. - post-validate strips hallucinated element refs (drops findings whose refs ALL fail to resolve; keeps findings with zero refs since some inconsistencies are genuinely about absences). apps/web/prisma/schema.prisma - Finding table: kind / text / linkedElementIds (JSON) / confidence / severity / validationCode / status / modelVersion / provider / model. - ResearchFinding table reserved for the Tavily integration that comes next — schema in place so we don't have to migrate again. apps/web/lib/db/repo.ts - listOpenFindings(projectId), replaceFindings(...) — replaceFindings wipes prior open findings in a transaction and writes the new set so re-detect doesn't accumulate stale findings. apps/web/app/api/projects/[projectId]/findings/route.ts - GET returns persisted open findings. - POST runs detect, persists, returns findings + meta (provider, model, durationMs, strippedRefs, droppedFindings). apps/web/components/editor/FindingsPanel.tsx - New panel, anchored bottom-right just left of IssuesPanel. Shows count summary (asm / risk / inc), detect / re-detect button, list grouped by kind (inconsistencies first, then risks, then assumptions), per-finding glyph + tag + severity + confidence + linked refs. - Click a finding row → focus its first linked element via the same setFocusBlockId path the rail and IssuesPanel already use. - "stale" indicator when the model version has advanced past the one the findings were detected against. EditorShell wires version + projectId through to FindingsPanel. Smoke-tested end-to-end: 12 findings returned (5 asm / 4 risk / 3 inc), 0 hallucinated refs stripped, ~37s on local gemma-4-e4b. Sample assumption "students are willing to engage with an AI tutor that is programmed to refuse providing complete solutions" — specific to Aristotle's refusal_policy, not a generic startup truism. Deferred to follow-ups: inline rail/diagram badges from findings, auto-detect-on-save, Tavily research, experiment modal.
This commit is contained in:
@@ -26,6 +26,7 @@ model Project {
|
||||
snapshots ModelSnapshot[]
|
||||
changes ChangelogEntry[]
|
||||
threads SocratesThread[]
|
||||
findings Finding[]
|
||||
}
|
||||
|
||||
model ModelSnapshot {
|
||||
@@ -87,3 +88,55 @@ model SocratesMessage {
|
||||
|
||||
@@index([threadId, createdAt])
|
||||
}
|
||||
|
||||
/// Background-detected finding (assumption / risk / inconsistency).
|
||||
/// Findings are scoped to the model version that produced them; on every
|
||||
/// detection run we delete the project's previous open findings and write
|
||||
/// the new set so we don't accumulate stale ones across model edits.
|
||||
model Finding {
|
||||
id String @id @default(cuid())
|
||||
projectId String
|
||||
/// "assumption" | "risk" | "inconsistency"
|
||||
kind String
|
||||
text String
|
||||
/// JSON-encoded string[] of element ids this finding references.
|
||||
linkedElementIds String
|
||||
confidence Float
|
||||
/// Risks only: "low" | "medium" | "high"
|
||||
severity String?
|
||||
/// Inconsistencies only: optional structural-rule code (S1, M2, T1, …)
|
||||
validationCode String?
|
||||
/// Lifecycle: "open" | "dismissed" | "resolved"
|
||||
status String @default("open")
|
||||
/// Model version this finding was detected against.
|
||||
modelVersion Int
|
||||
/// Provider + model that emitted the finding.
|
||||
provider String?
|
||||
llmModel String?
|
||||
createdAt DateTime @default(now())
|
||||
|
||||
project Project @relation(fields: [projectId], references: [id], onDelete: Cascade)
|
||||
research ResearchFinding[]
|
||||
|
||||
@@index([projectId, status])
|
||||
@@index([projectId, kind])
|
||||
}
|
||||
|
||||
/// Web research results from validating a finding via Tavily / similar.
|
||||
/// Deferred to a follow-up commit; schema is here so we don't have to
|
||||
/// migrate again later.
|
||||
model ResearchFinding {
|
||||
id String @id @default(cuid())
|
||||
findingId String
|
||||
query String
|
||||
url String
|
||||
title String?
|
||||
snippet String?
|
||||
/// "supports" | "contradicts" | "neutral"
|
||||
stance String @default("neutral")
|
||||
createdAt DateTime @default(now())
|
||||
|
||||
finding Finding @relation(fields: [findingId], references: [id], onDelete: Cascade)
|
||||
|
||||
@@index([findingId])
|
||||
}
|
||||
|
||||
Reference in New Issue
Block a user