@index9/mcp 6.1.0 → 6.2.0

package/dist/cli.js

@@ -48,6 +48,11 @@ var MissingModelDiagnosticSchema = z.object({
   provider: z.string().optional(),
   message: z.string()
 });
+var SuggestionEntrySchema = z.object({
+  id: z.string(),
+  name: z.string(),
+  created: z.number().nullable()
+});
 var UserContentTextPartSchema = z.strictObject({
   type: z.literal("text"),
   text: z.string().trim().min(1)
@@ -108,8 +113,9 @@ Key rules:
 - find_models price-asc tends to be dominated by free preview models \u2014 pass \`excludeFree=true\` when you want a paid SLA.
 - find_models always emits \`meta.confidence\` ("high" | "low") on semantic queries. Low means no candidate matched on keyword (BM25); \`meta.lowConfidenceReason\` is "no_keyword_matches" or "no_results" and \`meta.suggestion\` carries an actionable hint. Weak hits are capped at score=30 so they don't masquerade as strong matches. Pass \`requireKeywordMatch: true\` to get an empty page instead of weak vector-only neighbors.
 - find_models with sortBy=price exposes \`pricing.effectivePromptPerMillion\` and \`pageInfo.priceSortBasis\` \u2014 sort order may diverge from displayed promptPerMillion for models with per-request fees.
-- get_models accepts aliases (display names, short names) \u2014 not just full IDs. Unknown ids return in missingIds with \`suggestions\` (token-fuzzy or recency-anchored newest-from-provider) and \`missingDiagnostics\` keyed by id with \`reason\` ("unknown_provider" | "no_match" | "suggestions_available" | "ambiguous_alias") so retry strategy is explicit. Retry with one of the suggested ids.
+- Your training-data model IDs are routinely stale. get_models / compare_models / test_model all accept aliases (display names, short names) and return unknown ids in \`missingIds\` with \`suggestions[id]\` ordered newest-first (each entry: \`{id, name, created}\`, where \`created\` is unix seconds), plus \`missingDiagnostics[id].reason\` \u2208 {"unknown_provider", "no_match", "suggestions_available", "ambiguous_alias"}. **Default recovery: retry with \`suggestions[id][0].id\` \u2014 it's the newest viable replacement.** If suggestions is empty or reason="no_match"/"unknown_provider", fall back to \`find_models sortBy=created\` instead.
 - compare_models accepts the same alias formats as get_models. Use it instead of N parallel get_models calls when the user is comparing finalists.
+- test_model pre-flight resolves and filters unresolvable ids out of the OpenRouter call, so stale ids never cost you credits \u2014 they come back in missingIds with the same suggestions/diagnostics surface as get_models. If every id is unresolvable, the call returns 400 with diagnostics and no inference fires.
 - Use test_model with \`dryRun=true\` to estimate cost before live testing. Pass \`expectedPromptTokens\` for capacity planning at sizes you don't want to paste in full.
 - test_model with \`dryRun=false\` (default) requires OPENROUTER_API_KEY and incurs real usage costs.
 - Reasoning-capable models (capabilities includes "reasoning") burn hidden reasoning tokens against \`maxTokens\` before emitting visible text. Leave \`maxTokens\` unset, or set it to at least 2000, when testing reasoning models \u2014 otherwise results may fail with finish_reason=length.
@@ -152,7 +158,7 @@ Pass result.id to get_models for full specs or to test_model for live testing.`,
 
 Call after find_models to inspect candidates, or directly when the user names a model (format: 'provider/model-name').
 
-Response: { results: (Model | null)[], missingIds: string[], resolvedAliases?: Record<alias, canonicalId>, ambiguousAliases?: Record<alias, candidateIds[]>, suggestions?: Record<unknownId, candidateIds[]> }. Each non-null result has:
+Response: { results: (Model | null)[], missingIds: string[], resolvedAliases?: Record<alias, canonicalId>, ambiguousAliases?: Record<alias, candidateIds[]>, suggestions?: Record<unknownId, Array<{id, name, created}>> }. Each non-null result has:
 - id, canonicalSlug, name, description
 - created (unix seconds), createdAt (ISO 8601), knowledgeCutoff (ISO date or null)
 - contextLength (tokens), maxOutputTokens, isModerated
@@ -161,7 +167,7 @@ Response: { results: (Model | null)[], missingIds: string[], resolvedAliases?: R
 - capabilities[]: normalized capability flags (same values as find_models and capabilitiesAll/Any)
 - supportedParameters[]: OpenRouter parameters the model accepts (e.g., "temperature", "tools", "response_format")
 
-Entries in results are null when the id is unknown; those ids appear in missingIds. Ambiguous aliases appear in ambiguousAliases with candidate canonical ids \u2014 pass a canonical id to disambiguate. Unknown ids that partially match (e.g. "sonnet" \u2192 all Claude Sonnet variants) appear in suggestions with up to 5 candidate ids. When token-overlap finds nothing but the id is shaped like \`provider/<unknown>\` and the provider exists, suggestions falls back to the 5 newest models from that provider (real created timestamps, no hardcoded "popular" list). Retry with one of the suggested ids.
+Entries in results are null when the id is unknown; those ids appear in missingIds. Ambiguous aliases appear in ambiguousAliases with candidate canonical ids \u2014 pass a canonical id to disambiguate. Unknown ids that partially match (e.g. "sonnet" \u2192 all Claude Sonnet variants) appear in \`suggestions\` as up to 5 \`{id, name, created}\` entries **sorted newest-first** \u2014 pick \`suggestions[id][0].id\` for the most current replacement without a second lookup. When token-overlap finds nothing but the id is shaped like \`provider/<unknown>\` and the provider exists, suggestions falls back to the 5 newest models from that provider (real created timestamps, no hardcoded "popular" list).
 
 \`missingDiagnostics\` (when present) gives a machine-readable reason per missing id: \`unknown_provider\` (the prefix before / isn't in the catalog \u2014 fix the provider, not the model name), \`ambiguous_alias\`, \`suggestions_available\` (mirrors suggestions[id]), or \`no_match\`.`,
     requiresKey: false
@@ -174,13 +180,13 @@ Entries in results are null when the id is unknown; those ids appear in missingI
 
 Use this when the user asks "which is cheaper / has more context / supports X" across multiple specific models. Faster than calling get_models and diffing yourself.
 
-Response: { models: ModelResponse[], diff: { contextLength, maxOutputTokens, promptPricePerMillion, completionPricePerMillion, tokenizer, inputModalities, outputModalities, capabilities, supportedParameters }, cheapestForPromptPerMillion, largestContext, missingIds, resolvedAliases?, ambiguousAliases?, suggestions? }.
+Response: { models: ModelResponse[], diff: { contextLength, maxOutputTokens, promptPricePerMillion, completionPricePerMillion, tokenizer, inputModalities, outputModalities, capabilities, supportedParameters }, cheapestForPromptPerMillion, largestContext, missingIds, resolvedAliases?, ambiguousAliases?, suggestions?: Record<unknownId, Array<{id, name, created}>> (newest-first), missingDiagnostics? }.
 
 Each numeric/string diff field has { allEqual: boolean, values: Record<id, value|null> }. Capability/parameter diffs have { commonAll: string[], uniquePerModel: Record<id, string[]> }. cheapestForPromptPerMillion / largestContext are convenience picks across the supplied models \u2014 null when the field is missing on every model.
 
 Optional: pass \`expectedPromptTokens\` AND \`expectedCompletionTokens\` to also receive \`workloadCosts\` and \`cheapestForRealisticWorkload\` \u2014 the actual cheapest given the user's expected token mix. Each \`workloadCosts[i]\` carries \`tokenCostUsd\` (token-only), \`requestCostUsd\` (per-request fee), \`totalCostUsd\` (sum, includes request fees), and \`pricingBasis\` ("exact_per_token" | "rounded_per_million" | "unavailable"). This matters when prompt:completion price ratios diverge across models, or when a model has a per-request fee.
 
-Accepts the same alias formats as get_models. Unknown ids are returned in missingIds (with suggestions when partial matches exist, plus \`missingDiagnostics\` carrying a machine-readable reason per id).`,
+Accepts the same alias formats as get_models. Unknown ids are returned in missingIds (with \`suggestions[id]\` as newest-first \`{id, name, created}\` entries when partial matches exist, plus \`missingDiagnostics\` carrying a machine-readable reason per id). When fewer than 2 ids resolve, this returns 400 with the diagnostics so you can retry with \`suggestions[id][0].id\` for each missing id.`,
     requiresKey: false
   },
   list_facets: {
@@ -216,7 +222,9 @@ Parameters:
 
 Results (live): each result carries modelId (the id you passed), resolvedModelId (canonical id, present when the input was an alias), ok, response, latencyMs, tokens { prompt, completion }, cost (USD; live from OpenRouter when available, else estimated from cached pricing), and truncated=true when finish_reason is "length". On failure, results include \`error\` (free-form) plus \`failureReason\` ("insufficient_credits" | "model_unavailable" | "rate_limited" | "timeout" | "invalid_request" | "unknown") so callers can pick a retry strategy without parsing the error string.
 
-Results (dryRun): each entry carries \`tokenCostUsd\`, \`requestCostUsd\`, \`totalCostUsd\` (matches \`estimatedCost\`, includes per-request fees), and \`estimatedCostBasis\` (same enum as compare_models.workloadCosts). Use find_models or get_models first to identify model ids.`,
+Results (dryRun): each entry carries \`tokenCostUsd\`, \`requestCostUsd\`, \`totalCostUsd\` (matches \`estimatedCost\`, includes per-request fees), and \`estimatedCostBasis\` (same enum as compare_models.workloadCosts). Use find_models or get_models first to identify model ids.
+
+Stale-id recovery: unresolvable model ids are filtered out **before** any OpenRouter call (so they cost nothing) and returned in \`missingIds\` alongside \`suggestions\` (newest-first \`{id, name, created}\` entries), \`resolvedAliases\`, \`ambiguousAliases\`, and \`missingDiagnostics\` \u2014 same shape as get_models / compare_models. If every id is unresolvable, the call returns 400 with diagnostics and no inference fires. Default recovery: retry with \`suggestions[id][0].id\`.`,
     requiresKey: true
   }
 };
@@ -654,7 +662,7 @@ var BatchModelLookupResponseSchema = z3.object({
   missingIds: z3.array(z3.string()),
   resolvedAliases: z3.record(z3.string(), z3.string()).optional(),
   ambiguousAliases: z3.record(z3.string(), z3.array(z3.string())).optional(),
-  suggestions: z3.record(z3.string(), z3.array(z3.string())).optional(),
+  suggestions: z3.record(z3.string(), z3.array(SuggestionEntrySchema)).optional(),
   missingDiagnostics: z3.record(z3.string(), MissingModelDiagnosticSchema).optional()
 }).strict();
 var GetModelsToolResultSchema = z3.object({
@@ -662,7 +670,7 @@ var GetModelsToolResultSchema = z3.object({
   missingIds: z3.array(z3.string()),
   resolvedAliases: z3.record(z3.string(), z3.string()).optional(),
   ambiguousAliases: z3.record(z3.string(), z3.array(z3.string())).optional(),
-  suggestions: z3.record(z3.string(), z3.array(z3.string())).optional(),
+  suggestions: z3.record(z3.string(), z3.array(SuggestionEntrySchema)).optional(),
   missingDiagnostics: z3.record(z3.string(), MissingModelDiagnosticSchema).optional(),
   _index9: Index9MetaSchema
 });
@@ -720,7 +728,7 @@ var CompareResponseSchema = z4.object({
   workloadCosts: z4.array(CompareWorkloadCostSchema).optional(),
   resolvedAliases: z4.record(z4.string(), z4.string()).optional(),
   missingIds: z4.array(z4.string()),
-  suggestions: z4.record(z4.string(), z4.array(z4.string())).optional(),
+  suggestions: z4.record(z4.string(), z4.array(SuggestionEntrySchema)).optional(),
   ambiguousAliases: z4.record(z4.string(), z4.array(z4.string())).optional(),
   missingDiagnostics: z4.record(z4.string(), MissingModelDiagnosticSchema).optional()
 }).strict();
@@ -850,13 +858,22 @@ var TestEstimateResultSchema = z6.object({
   totalCostUsd: z6.number().nullable().optional(),
   estimatedCostBasis: PricingBasisSchema.optional()
 });
+var TestResolutionFieldsSchema = {
+  missingIds: z6.array(z6.string()).optional(),
+  resolvedAliases: z6.record(z6.string(), z6.string()).optional(),
+  ambiguousAliases: z6.record(z6.string(), z6.array(z6.string())).optional(),
+  suggestions: z6.record(z6.string(), z6.array(SuggestionEntrySchema)).optional(),
+  missingDiagnostics: z6.record(z6.string(), MissingModelDiagnosticSchema).optional()
+};
 var TestDryRunResponseSchema = z6.object({
   dryRun: z6.literal(true),
   results: z6.array(TestEstimateResultSchema),
-  disclaimer: z6.string()
+  disclaimer: z6.string(),
+  ...TestResolutionFieldsSchema
 });
 var TestLiveResponseSchema = z6.object({
-  results: z6.array(TestResultSchema)
+  results: z6.array(TestResultSchema),
+  ...TestResolutionFieldsSchema
 });
 var TestResponseSchema = z6.union([TestDryRunResponseSchema, TestLiveResponseSchema]);
 
@@ -1004,6 +1021,22 @@ function extractError(body) {
   }
   return "Request failed";
 }
+var RECOVERY_FIELDS = [
+  "missingIds",
+  "resolvedAliases",
+  "ambiguousAliases",
+  "suggestions",
+  "missingDiagnostics"
+];
+function extractRecoveryFields(body) {
+  if (typeof body !== "object" || body === null || Array.isArray(body)) return {};
+  const out = {};
+  const b = body;
+  for (const key of RECOVERY_FIELDS) {
+    if (key in b) out[key] = b[key];
+  }
+  return out;
+}
 async function callApi(ctx, url, options, responseSchema) {
   const res = await fetchWithRetry(url, options);
   let body;
@@ -1014,7 +1047,12 @@ async function callApi(ctx, url, options, responseSchema) {
   }
   if (!res.ok) {
     return toResponse(
-      { error: extractError(body), status: res.status, _index9: buildMeta(ctx, res.headers) },
+      {
+        error: extractError(body),
+        status: res.status,
+        ...extractRecoveryFields(body),
+        _index9: buildMeta(ctx, res.headers)
+      },
       true
     );
   }
@@ -1062,7 +1100,11 @@ async function handleGetModels(ctx, args) {
   return callApi(
     ctx,
     `${ctx.baseUrl}${API_PATHS.model}`,
-    { method: "POST", headers: baseHeaders(ctx), body: JSON.stringify(parsed.data) },
+    {
+      method: "POST",
+      headers: baseHeaders(ctx),
+      body: JSON.stringify(parsed.data)
+    },
     BatchModelLookupResponseSchema
   );
 }
@@ -1074,7 +1116,11 @@ async function handleCompareModels(ctx, args) {
   return callApi(
     ctx,
     `${ctx.baseUrl}${API_PATHS.compare}`,
-    { method: "POST", headers: baseHeaders(ctx), body: JSON.stringify(parsed.data) },
+    {
+      method: "POST",
+      headers: baseHeaders(ctx),
+      body: JSON.stringify(parsed.data)
+    },
     CompareResponseSchema
   );
 }

package/manifest.json

@@ -1,7 +1,7 @@
 {
   "manifest_version": "0.3",
   "name": "index9",
-  "version": "6.0.0",
+  "version": "6.1.0",
   "description": "Discover, shortlist, compare, cost-model, and live-test 300+ AI models from your editor",
   "author": {
     "name": "Index9"

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@index9/mcp",
-  "version": "6.1.0",
+  "version": "6.2.0",
   "license": "MIT",
   "repository": {
     "type": "git",
@@ -24,11 +24,11 @@
     "zod": "^4.4.3"
   },
   "devDependencies": {
-    "@types/node": "^25.6.1",
+    "@types/node": "^25.6.2",
     "tsup": "^8.5.1",
     "typescript": "6.0.3",
-    "vitest": "^4.1.5",
-    "@index9/core": "2.4.0"
+    "vitest": "^4.1.6",
+    "@index9/core": "2.5.0"
   },
   "engines": {
     "node": ">=20"