fix(confidence-engine): unwrap overview provider response

Production: synthesizeInvestigationOverview() now passes providerResult.response
to validateOverviewResponse() instead of the raw envelope, matching the proven
pattern in current-understanding-synthesis.js.

Regression: fakeProvider returns real-provider envelope shape { response, providerApiPath }
and four new regression tests verify unwrapping succeeds with valid response
and fails correctly with invalid response inside the envelope.

Handoff: documented defect confirmed, fix applied, and owning regression added.
This commit is contained in:
2026-09-30 16:06:14 +01:00
parent 8d268efb12
commit 3d0c90e64c
3 changed files with 76 additions and 8 deletions
+8
View File
@@ -385,6 +385,14 @@ RAW USER EVIDENCE
- Current Understanding and Investigation Report are architecturally distinct artefacts. Report interpretations remain explicit, not evidence.
- The `≤5` processing bound was experimental apparatus constraint, NOT product requirement.
## Investigation overview synthesis — provider envelope defect confirmed
- `synthesizeInvestigationOverview()` previously passed the full provider envelope (`{ response: { understanding, plausibleInterpretations }, providerApiPath, providerExecution }`) directly into `validateOverviewResponse()`, which expected only `{ understanding, plausibleInterpretations }`. This caused "Missing or invalid required fields" on real providers.
- Fix now unwraps `.response` before overview validation, matching the proven pattern in `lib/graph/current-understanding-synthesis.js` (line 296: `validateSynthesisResponse(providerResult.response)`).
- Owning regression (`tests/graph/investigation-overview-synthesis.test.js`) now mirrors real provider envelope shape (`{ response: { understanding, plausibleInterpretations }, providerApiPath }`). Regression fails against pre-fix seam and passes after fix.
- No live proof yet. Zero live calls made in this handoff.
- Next boundary: one bounded live product-launch overview Call A before resuming A/B epistemic input control comparison.
## HISTORICAL — initial-decomposition v0.61 status
Initial decomposition experiments (v0.61) ran four successful runs plus one validation failure against the same manufacturing scenario. Key findings:
@@ -240,9 +240,9 @@ export async function synthesizeInvestigationOverview(params, deps) {
// Build overview-specific prompt (keeps evidence/interpretation separate)
const prompt = buildOverviewSynthesisPrompt(situationGraph, findings, plausibleInterpretations);
let rawResponse;
let providerResult;
try {
rawResponse = await provider.generateReconstruction(
providerResult = await provider.generateReconstruction(
prompt,
modelName,
);
@@ -250,8 +250,8 @@ export async function synthesizeInvestigationOverview(params, deps) {
throw new Error(`Overview synthesis provider call failed: ${err.message}`);
}
// Validate against overview-specific contract
const validated = validateOverviewResponse(rawResponse);
// Validate against overview-specific contract — unwrap provider envelope
const validated = validateOverviewResponse(providerResult.response);
if (!validated.valid) {
throw new Error(`Overview synthesis validation failed: ${validated.reason}`);
}
@@ -39,10 +39,13 @@ const fakeProvider = (defaultResponse) => ({
generateReconstruction: vi.fn(() => {
return typeof defaultResponse === "function"
? defaultResponse()
: JSON.stringify({
understanding: "Synthesized understanding from evidence.",
plausibleInterpretations: "Remaining plausible explanations.",
});
: {
response: {
understanding: "Synthesized understanding from evidence.",
plausibleInterpretations: "Remaining plausible explanations.",
},
providerApiPath: "/api/chat",
};
}),
});
@@ -459,4 +462,61 @@ describe("synthesizeInvestigationOverview — full seam", () => {
expect(eligible.map((f) => f.id)).toEqual(["f1", "f2"]);
expect(eligible).toHaveLength(2);
});
// ── Regression: overview synthesis unwraps provider envelope before validation ───────────────
describe("overview synthesis — provider envelope unwrapping regression", () => {
it("fake provider returns real-provider envelope shape (response + providerApiPath)", async () => {
const fake = fakeProvider();
await synthesizeInvestigationOverview(
{ situationGraph: graph, findings: [], plausibleInterpretations: [] },
{ provider: fake }
);
const callResult = fake.generateReconstruction.mock.calls[0][2];
expect(fake.generateReconstruction).toHaveBeenCalledTimes(1);
});
it("valid envelope response is unwrapped and validated successfully", async () => {
const fake = fakeProvider(() => ({
response: JSON.stringify({
understanding: "Test understanding.",
plausibleInterpretations: "Test interpretations.",
}),
providerApiPath: "/v1/responses",
}));
const result = await synthesizeInvestigationOverview(
{ situationGraph: graph, findings: [], plausibleInterpretations: [] },
{ provider: fake }
);
expect(result.understanding).toBe("Test understanding.");
expect(result.plausibleInterpretations).toBe("Test interpretations.");
});
it("invalid response inside envelope fails validation (regression guard)", async () => {
const fake = fakeProvider(() => ({
response: JSON.stringify({ wrongField: "value" }),
providerApiPath: "/v1/responses",
}));
await expect(
synthesizeInvestigationOverview(
{ situationGraph: graph, findings: [], plausibleInterpretations: [] },
{ provider: fake }
)
).rejects.toThrow(/validation failed/);
});
it("provider returns .response as raw valid JSON string inside envelope", async () => {
const json = JSON.stringify({
understanding: "Raw JSON test.",
plausibleInterpretations: "String payload.",
});
const fake = fakeProvider(() => ({ response: json, providerApiPath: "/api/chat" }));
const result = await synthesizeInvestigationOverview(
{ situationGraph: graph, findings: [], plausibleInterpretations: [] },
{ provider: fake }
);
expect(result.understanding).toBe("Raw JSON test.");
expect(fake.generateReconstruction).toHaveBeenCalledTimes(1);
});
});
});