@sellable/mcp 0.1.104 → 0.1.106

package/agents/source-scout-linkedin-engagement.md

@@ -7,10 +7,10 @@ Required first step:
 - Load the canonical provider prompt before searching. If the parent supplies a
   draft `campaignOfferId`, call `get_provider_prompt({ provider:
 "signal-discovery", campaignOfferId, confirmed: true })` and include that same
-  `campaignOfferId` in `search_signals` so the user can watch source work in the
-  campaign UI. Treat that as a campaign-attached persisted search; do not run a
-  post-mint search without the campaign ID. If no campaign ID is supplied, run
-  campaignless preview mode.
+  `campaignOfferId` plus `currentStep: "signal-discovery"` in `search_signals`
+  so the user can watch source work in the campaign UI. Treat that as a
+  campaign-attached persisted search; do not run a post-mint search without the
+  campaign ID. If no campaign ID is supplied, run campaignless preview mode.
 
 Use the inherited Sellable MCP tools when available:
 

package/agents/source-scout-prospeo-contact.md

@@ -6,9 +6,10 @@ Required first step:
 
 - Load the canonical provider prompt before searching. If the parent supplies a
   draft `campaignOfferId`, call `get_provider_prompt({ provider: "prospeo",
-campaignOfferId, confirmed: true })` and include that same `campaignOfferId` in
-  `search_prospeo` so the user can watch source work in the campaign UI. If no
-  campaign ID is supplied, run campaignless preview mode. Treat post-mint
+campaignOfferId, confirmed: true })` and include that same `campaignOfferId`
+  plus `currentStep: "prospeo"` in `search_prospeo` so the user can watch source
+  work in the campaign UI. If no campaign ID is supplied, run campaignless
+  preview mode. Treat post-mint
   searches with `campaignOfferId` as campaign-attached persisted search tabs;
   do not run a live campaign search without the campaign ID.
 

package/agents/source-scout-sales-nav.md

@@ -7,8 +7,9 @@ Required first step:
 - Load the canonical provider prompt before searching. If the parent supplies a
   draft `campaignOfferId`, call `get_provider_prompt({ provider: "sales-nav",
 campaignOfferId, confirmed: true })` and include that same `campaignOfferId` in
-  `search_sales_nav` so the user can watch source work in the campaign UI. If no
-  campaign ID is supplied, run campaignless preview mode. Treat post-mint
+  `search_sales_nav` with `currentStep: "sales-nav"` so the user can watch
+  source work in the campaign UI. If no campaign ID is supplied, run
+  campaignless preview mode. Treat post-mint
   searches with `campaignOfferId` as campaign-attached persisted search tabs;
   do not run a live campaign search without the campaign ID.
 

package/dist/tools/leads.js

@@ -594,7 +594,7 @@ export const leadToolDefinitions = [
     },
     {
         name: "search_sales_nav",
-        description: 'Search LinkedIn Sales Navigator. Requires get_provider_prompt({ provider: "sales-nav" }) first. Returns normalized results with pagination. `campaignOfferId` is optional for directional preview runs; include it only when the search should be associated with a campaign.',
+        description: 'Search LinkedIn Sales Navigator. Requires get_provider_prompt({ provider: "sales-nav" }) first. Returns normalized results with pagination. `campaignOfferId` is optional for directional preview runs; include it only when the search should be associated with a campaign. Post-mint create-campaign-v2 watch runs MUST pass campaignOfferId and currentStep: "sales-nav" so the search appears in the watched campaign UI. Omitting campaignOfferId post-mint orphans the search from the UI.',
         inputSchema: {
             type: "object",
             properties: {
@@ -739,7 +739,7 @@ export const leadToolDefinitions = [
     },
     {
         name: "search_prospeo",
-        description: 'Search Prospeo for people using filters and optional domainFilterId. Requires get_provider_prompt({ provider: "prospeo" }) first. When targeting known accounts, call load_csv_domains for CSV-on-disk workflows or save_domain_filters for pasted/raw domain lists, then pass domainFilterId. Raw domain inputs and company-name targeting are NOT supported in this MCP tool. Strategy: start with 2-3 high-signal filters (title/seniority + industry or domainFilterId + headcount), then tighten one filter at a time. For security, AppSec, SOC, RevOps, Demand Gen, and similar function-specific lanes, do not widen with bare seniority labels like "Head" or "Director" alone; pair them with explicit function-title keywords and inspect the sample for off-function `Head of X` leakage. Prefer person location over company HQ unless HQ is explicitly needed. `campaignOfferId` routing rule: OMIT campaignOfferId ONLY in pre-mint Phase 84 `find leads` discovery mode (validating ICP before the commit gate). In every other context — post-mint lead additions, operator-driven searches on a live campaign, any search where the intent is to persist results to a specific campaign — you MUST pass campaignOfferId so the search shows up in that campaign\'s Contact Search panel. Omitting campaignOfferId post-mint orphans the search from the UI. Returns normalized results with pagination.',
+        description: 'Search Prospeo for people using filters and optional domainFilterId. Requires get_provider_prompt({ provider: "prospeo" }) first. When targeting known accounts, call load_csv_domains for CSV-on-disk workflows or save_domain_filters for pasted/raw domain lists, then pass domainFilterId. Raw domain inputs and company-name targeting are NOT supported in this MCP tool. Strategy: start with 2-3 high-signal filters (title/seniority + industry or domainFilterId + headcount), then tighten one filter at a time. For security, AppSec, SOC, RevOps, Demand Gen, and similar function-specific lanes, do not widen with bare seniority labels like "Head" or "Director" alone; pair them with explicit function-title keywords and inspect the sample for off-function `Head of X` leakage. Prefer person location over company HQ unless HQ is explicitly needed. `campaignOfferId` routing rule: OMIT campaignOfferId ONLY in pre-mint Phase 84 `find leads` discovery mode (validating ICP before the commit gate). In every other context — post-mint lead additions, operator-driven searches on a live campaign, any search where the intent is to persist results to a specific campaign — you MUST pass campaignOfferId so the search shows up in that campaign\'s Contact Search panel. Post-mint create-campaign-v2 watch runs MUST pass currentStep: "prospeo". Omitting campaignOfferId post-mint orphans the search from the UI. Returns normalized results with pagination.',
         inputSchema: {
             type: "object",
             properties: {
@@ -848,7 +848,7 @@ export const leadToolDefinitions = [
     },
     {
         name: "search_signals",
-        description: 'Search LinkedIn posts for signal discovery. Requires get_provider_prompt({ provider: "signal-discovery" }) first. Supports keyword search, profile posts, company posts, or a single post URL. Use campaignless preview mode for directional discovery; include `campaignOfferId` only when the search should persist into campaign state. Returns a compact summary with recommended posts to avoid context bloat.',
+        description: 'Search LinkedIn posts for signal discovery. Requires get_provider_prompt({ provider: "signal-discovery" }) first. Supports keyword search, profile posts, company posts, or a single post URL. Use campaignless preview mode for directional discovery; include `campaignOfferId` when the search should persist into campaign state. Post-mint create-campaign-v2 watch runs MUST pass campaignOfferId and currentStep: "signal-discovery" so the search appears in the watched campaign UI. Omitting campaignOfferId post-mint orphans the search from the UI. Returns a compact summary with recommended posts to avoid context bloat.',
         inputSchema: {
             type: "object",
             properties: {
@@ -2366,6 +2366,15 @@ export async function selectPromisingPosts(input) {
         headlineICPCriteria,
         rubricGuidelines: headlineICPCriteria,
     });
+    if (selectionResult.selectedCount <= 0) {
+        return {
+            success: false,
+            selectedCount: selectionResult.selectedCount,
+            unselectedCount: selectionResult.unselectedCount,
+            criteriaCount: selectionResult.criteriaCount,
+            message: "No Signal Discovery posts were selected for this campaign. Do not import yet. Re-run a campaign-scoped search_signals call, use recommendedPostIds from the current campaign search state, or ask the user to promote posts in the UI before retrying select_promising_posts.",
+        };
+    }
     // Update currentStep if provided (via v2 endpoint)
     if (currentStep) {
         await api.put(`/api/v2/campaign-offers/${campaignOfferId}`, {

package/dist/tools/rubrics.d.ts

@@ -32,6 +32,7 @@ type WaitForRubricResultsInput = {
     campaignOfferId?: string;
     tableId?: string;
     targetCount?: number;
+    minPassedCount?: number;
     timeoutMs?: number;
     intervalMs?: number;
     includeRows?: boolean;
@@ -109,6 +110,7 @@ export declare const rubricToolDefinitions: ({
             maxProspects?: undefined;
             tableId?: undefined;
             targetCount?: undefined;
+            minPassedCount?: undefined;
             timeoutMs?: undefined;
             intervalMs?: undefined;
             includeRows?: undefined;
@@ -141,6 +143,7 @@ export declare const rubricToolDefinitions: ({
             maxProspects?: undefined;
             tableId?: undefined;
             targetCount?: undefined;
+            minPassedCount?: undefined;
             timeoutMs?: undefined;
             intervalMs?: undefined;
             includeRows?: undefined;
@@ -198,6 +201,7 @@ export declare const rubricToolDefinitions: ({
             maxProspects?: undefined;
             tableId?: undefined;
             targetCount?: undefined;
+            minPassedCount?: undefined;
             timeoutMs?: undefined;
             intervalMs?: undefined;
             includeRows?: undefined;
@@ -255,6 +259,7 @@ export declare const rubricToolDefinitions: ({
             maxProspects?: undefined;
             tableId?: undefined;
             targetCount?: undefined;
+            minPassedCount?: undefined;
             timeoutMs?: undefined;
             intervalMs?: undefined;
             includeRows?: undefined;
@@ -314,6 +319,7 @@ export declare const rubricToolDefinitions: ({
             maxProspects?: undefined;
             tableId?: undefined;
             targetCount?: undefined;
+            minPassedCount?: undefined;
             timeoutMs?: undefined;
             intervalMs?: undefined;
             includeRows?: undefined;
@@ -343,6 +349,7 @@ export declare const rubricToolDefinitions: ({
             maxProspects?: undefined;
             tableId?: undefined;
             targetCount?: undefined;
+            minPassedCount?: undefined;
             timeoutMs?: undefined;
             intervalMs?: undefined;
             includeRows?: undefined;
@@ -372,6 +379,7 @@ export declare const rubricToolDefinitions: ({
             checkName?: undefined;
             tableId?: undefined;
             targetCount?: undefined;
+            minPassedCount?: undefined;
             timeoutMs?: undefined;
             intervalMs?: undefined;
             includeRows?: undefined;
@@ -397,6 +405,10 @@ export declare const rubricToolDefinitions: ({
                 type: string;
                 description: string;
             };
+            minPassedCount: {
+                type: string;
+                description: string;
+            };
             timeoutMs: {
                 type: string;
                 description: string;
@@ -496,6 +508,7 @@ export declare function waitForRubricResults(input: WaitForRubricResultsInput):
     elapsedMs: number;
     tableId: string;
     passRate: {
+        minPassedCount?: number | undefined;
         completed: number;
         passed: number;
         percent: number;
@@ -507,7 +520,10 @@ export declare function waitForRubricResults(input: WaitForRubricResultsInput):
     diagnostic?: undefined;
     guidance?: undefined;
 } | {
+    stats: WorkflowTableStats;
+    rows?: import("./rows.js").LightweightRow[] | undefined;
     ready: boolean;
+    partial: boolean;
     reason: string;
     attempts: number;
     elapsedMs: number;
@@ -518,6 +534,7 @@ export declare function waitForRubricResults(input: WaitForRubricResultsInput):
         pending: number;
         percent: number;
         targetCount: number;
+        minPassedCount: number;
     };
     partialResult: {
         completed: number;
@@ -525,7 +542,35 @@ export declare function waitForRubricResults(input: WaitForRubricResultsInput):
         pending: number;
         percent: number;
         targetCount: number;
+        minPassedCount: number;
+        enoughToDiagnose: boolean;
+        floorMet: boolean;
+    };
+    diagnostic?: undefined;
+    guidance?: undefined;
+} | {
+    ready: boolean;
+    reason: string;
+    attempts: number;
+    elapsedMs: number;
+    tableId: string;
+    passRate: {
+        minPassedCount?: number | undefined;
+        completed: number;
+        passed: number;
+        pending: number;
+        percent: number;
+        targetCount: number;
+    };
+    partialResult: {
         enoughToDiagnose: boolean;
+        floorMet: boolean;
+        minPassedCount?: number | undefined;
+        completed: number;
+        passed: number;
+        pending: number;
+        percent: number;
+        targetCount: number;
     };
     diagnostic: {
         totalRows: number;

package/dist/tools/rubrics.js

@@ -103,6 +103,11 @@ function resolveMaxProspects(value) {
     const parsed = typeof value === "number" && !Number.isNaN(value) ? value : fallback;
     return Math.min(MAX_PROSPECTS_LIMIT, Math.max(1, Math.floor(parsed)));
 }
+function resolveMinPassedCount(value) {
+    if (typeof value !== "number" || Number.isNaN(value))
+        return null;
+    return Math.min(MAX_PROSPECTS_LIMIT, Math.max(1, Math.floor(value)));
+}
 function mergeRubricsWithExisting(draftRubrics, existingRubrics) {
     const existingByCheckName = new Map();
     existingRubrics.forEach((item) => {
@@ -348,6 +353,10 @@ export const rubricToolDefinitions = [
                     type: "number",
                     description: `Number of completed rubric results to wait for (default ${DEFAULT_MAX_PROSPECTS}).`,
                 },
+                minPassedCount: {
+                    type: "number",
+                    description: "Optional pass floor for bounded create-campaign samples. When this floor is met and remaining target rows are resolved as failed/not processing, the tool returns ready:true with partial:true instead of forcing a timeout.",
+                },
                 timeoutMs: {
                     type: "number",
                     description: `Max time to wait in ms (default ${DEFAULT_TIMEOUT_MS}).`,
@@ -625,6 +634,7 @@ export async function waitForRubricResults(input) {
     const timeoutMs = Math.max(5000, input.timeoutMs ?? DEFAULT_TIMEOUT_MS);
     const intervalMs = Math.max(500, input.intervalMs ?? DEFAULT_INTERVAL_MS);
     const targetCount = resolveMaxProspects(input.targetCount);
+    const minPassedCount = resolveMinPassedCount(input.minPassedCount);
     const includeRows = input.includeRows !== false;
     let tableId = input.tableId;
     if (!tableId && input.campaignOfferId) {
@@ -642,10 +652,17 @@ export async function waitForRubricResults(input) {
         const stats = await api.get(`/api/v3/workflow-tables/${tableId}/stats`);
         lastStats = stats;
         const completed = stats.passRate?.completed ?? 0;
+        const passed = stats.passRate?.passed ?? 0;
         const totalRows = stats.totalRows ?? 0;
         const effectiveTarget = totalRows > 0 ? Math.min(targetCount, totalRows) : targetCount;
+        const pending = Math.max(effectiveTarget - completed, 0);
+        const failedCount = stats.failedCount ?? 0;
+        const processingCount = stats.processingCount ?? 0;
+        const needsEnrichCount = stats.needsEnrichCount ?? 0;
+        const minPassFloorMet = minPassedCount !== null && passed >= minPassedCount;
+        const unresolvedRowsResolvedAsFailures = pending > 0 && completed + failedCount >= effectiveTarget;
+        const noActiveProcessing = processingCount === 0 && needsEnrichCount === 0;
         if (completed >= effectiveTarget) {
-            const passed = stats.passRate?.passed ?? 0;
             const percent = completed > 0 ? Math.round((passed / completed) * 100) : 0;
             const rowSnapshot = includeRows
                 ? await getTableRowsMinimal(tableId, {
@@ -662,6 +679,45 @@ export async function waitForRubricResults(input) {
                     passed,
                     percent,
                     targetCount: effectiveTarget,
+                    ...(minPassedCount !== null ? { minPassedCount } : {}),
+                },
+                ...(rowSnapshot ? { rows: rowSnapshot.rows } : {}),
+                stats,
+            };
+        }
+        if (minPassFloorMet &&
+            unresolvedRowsResolvedAsFailures &&
+            noActiveProcessing) {
+            const percent = completed > 0 ? Math.round((passed / completed) * 100) : 0;
+            const rowSnapshot = includeRows
+                ? await getTableRowsMinimal(tableId, {
+                    limit: effectiveTarget,
+                })
+                : null;
+            return {
+                ready: true,
+                partial: true,
+                reason: "min_passed_count_met_with_resolved_failures",
+                attempts,
+                elapsedMs: Date.now() - start,
+                tableId,
+                passRate: {
+                    completed,
+                    passed,
+                    pending,
+                    percent,
+                    targetCount: effectiveTarget,
+                    minPassedCount,
+                },
+                partialResult: {
+                    completed,
+                    passed,
+                    pending,
+                    percent,
+                    targetCount: effectiveTarget,
+                    minPassedCount,
+                    enoughToDiagnose: true,
+                    floorMet: true,
                 },
                 ...(rowSnapshot ? { rows: rowSnapshot.rows } : {}),
                 stats,
@@ -687,6 +743,7 @@ export async function waitForRubricResults(input) {
             pending,
             percent,
             targetCount: effectiveTarget,
+            ...(minPassedCount !== null ? { minPassedCount } : {}),
         },
         partialResult: {
             completed,
@@ -694,7 +751,9 @@ export async function waitForRubricResults(input) {
             pending,
             percent,
             targetCount: effectiveTarget,
+            ...(minPassedCount !== null ? { minPassedCount } : {}),
             enoughToDiagnose: completed > 0,
+            floorMet: minPassedCount !== null && passed >= minPassedCount,
         },
         diagnostic: {
             totalRows,

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@sellable/mcp",
-  "version": "0.1.104",
+  "version": "0.1.106",
   "type": "module",
   "description": "Sellable MCP server for Claude Code and Codex campaign workflows",
   "main": "dist/index.js",

package/skills/create-campaign-v2/SKILL.md

@@ -432,6 +432,13 @@ to be worth a LinkedIn test. I'll compare source paths by expected volume,
 sampled ICP fit, activity/warmth signal, cleanup risk, and tradeoffs. This
 usually takes ~3-5 min, and I'll show you the source decision + sample before
 anything goes live.`
+- In watch mode, do not leave the user sitting on only `pick-provider` while
+  source scouts run. Move the campaign to the likely primary source lane
+  (`signal-discovery`, `sales-nav`, or `prospeo`) before background source
+  scouts, then run the first campaign-attached provider prompt + provider search
+  in the parent thread with `campaignOfferId`, `confirmed: true`, and
+  `currentStep` when the tool accepts it. Background scouts may continue after
+  the visible campaign tab exists.
 - After the lead sample/source decision is ready and approved,
   show the next progress line:
   `Lead source is set. I'll import the first 15-row review batch into the
@@ -542,6 +549,11 @@ message we should test.`
 then draft the message from the same sample.` Never say `kicking off two
 workstreams`, `in parallel`, or `background` unless parallel branches were
   actually launched.
+- Source scout parallelism must not hide all provider work from the watch UI.
+  The parent thread owns the first visible provider search: call the chosen
+  provider prompt/search with the minted CampaignOffer id and provider
+  `currentStep` before waiting on source-scout results, so relevant search tabs
+  appear in real time.
 - For post-lead workstreams, first call
   `get_post_find_leads_scout_registry` and launch exactly the returned
   `filter-leads` and `message-generation` scouts when real subagents are

package/skills/create-campaign-v2/core/flow.v2.json

@@ -520,6 +520,33 @@
           "when": "before_source_scouts_or_provider_search",
           "chatRenderRule": "Move the campaign watch view to Find Contacts before the main thread starts comparing source paths. Do not mention MCP or local artifacts."
         },
+        {
+          "action": "advance_watch_to_initial_source_lane",
+          "tool": "update_campaign",
+          "requiredFields": [
+            "campaignId",
+            "currentStep"
+          ],
+          "currentStepByPrimaryLane": {
+            "signals": "signal-discovery",
+            "salesNav": "sales-nav",
+            "prospeo": "prospeo",
+            "apollo": "apollo-ai",
+            "existingList": "saved-lists",
+            "uploadedDomains": "prospeo"
+          },
+          "when": "before_background_source_scouts",
+          "rule": "Choose the likely primary visible source lane from source intake, brief preference, or the best first lane the main thread will actually search. Do this before waiting on background scouts so watch mode never sits on only Pick Provider while source work happens elsewhere."
+        },
+        {
+          "action": "run_first_campaign_attached_source_search",
+          "requiredFields": [
+            "campaignOfferId",
+            "currentStep"
+          ],
+          "before": "waiting_on_background_source_scouts",
+          "rule": "Before relying on background source scouts, the parent thread must run the first provider prompt + provider search for the chosen visible lane with campaignOfferId and currentStep (signal-discovery, sales-nav, or prospeo) so the campaign UI creates the corresponding live search/tab. Background scouts may continue after that, but the user must see at least one relevant campaign-attached search begin in the watched page."
+        },
         {
           "action": "render_find_leads_progress",
           "requiredVisibleContent": [
@@ -555,7 +582,7 @@
           "action": "run_subskill",
           "target": "find-leads",
           "mode": "campaign-attached-required",
-          "sourceScoutRule": "Shell-first flow requires the CampaignOffer campaignId from durable state. Pass campaignId as campaignOfferId into every provider prompt/search that can persist source state (`get_provider_prompt({ provider, campaignOfferId, confirmed: true })`, `search_signals`, `search_sales_nav`, `search_prospeo`) so the user can watch the selected source inside the campaign. The later import_leads call must use the same campaignOfferId. When several source lanes are credible, scout them independently, then write one primary source recommendation and any runner-up tradeoffs. Do not import, confirm, enrich, queue, or start leads during source discovery."
+          "sourceScoutRule": "Shell-first flow requires the CampaignOffer campaignId from durable state. Pass campaignId as campaignOfferId into every provider prompt/search that can persist source state (`get_provider_prompt({ provider, campaignOfferId, confirmed: true })`, `search_signals`, `search_sales_nav`, `search_prospeo`) and include currentStep for tools that accept it so the user can watch the selected source inside the campaign. In watch mode, run the first campaign-attached provider search in the parent thread before waiting on background scouts; subagents can explore runner-up lanes after the visible campaign tab exists. The later import_leads call must use the same campaignOfferId. When several source lanes are credible, scout them independently, then write one primary source recommendation and any runner-up tradeoffs. Do not import, confirm, enrich, queue, or start leads during source discovery."
         },
         {
           "action": "optional_debug_artifacts",
@@ -1916,17 +1943,23 @@
         {
           "tool": "wait_for_rubric_results",
           "requiredFields": [
-            "targetCount"
+            "targetCount",
+            "minPassedCount"
           ],
           "targetCountSource": "stats.totalRows_or_imported_batch_count",
+          "minPassedCountSource": "sample.minProjectedPass (3)",
           "requiredValues": {
             "includeRows": false
           },
-          "note": "The shell-first flow tests 15 leads first; always pass cohortSize explicitly instead of relying on default 25 behavior.",
+          "note": "The shell-first flow tests 15 leads first; always pass cohortSize explicitly instead of relying on default 25 behavior. Pass minPassedCount so a 15-row batch with resolved failures can advance when the 3-pass floor is already met.",
           "readVia": "stats_only_tool_result",
           "extractFields": [
             "ready",
+            "partial",
+            "reason",
             "passRate.completed",
+            "passRate.passed",
+            "passRate.minPassedCount",
             "stats"
           ],
           "doNotRetain": "rows_payload",

package/skills/create-campaign-v2-tail/SKILL.md

@@ -224,6 +224,17 @@ campaignOfferId, confirmed: true })` -> `search_signals({ campaignOfferId,
 selectionMode: "replace", selections, headlineICPCriteria })` ->
      `import_leads({ campaignOfferId, provider: "signal-discovery",
 targetLeadCount: importLimit })`.
+     For Signal Discovery, the promotion/select step is load-bearing. Use
+     post IDs from the current campaign-scoped `search_signals` response or
+     posts the user has visibly promoted in the campaign UI. Never use post IDs
+     copied only from a source-scout summary unless they have been re-resolved
+     through the current campaign search state. After `select_promising_posts`,
+     require `selectedCount > 0` before calling `import_leads`. If it returns
+     `selectedCount: 0`, do not switch providers and do not retry import.
+     Explain that the campaign has no promoted Signal Discovery posts yet,
+     re-run a narrow campaign-scoped `search_signals` call to recover current
+     post rows, or ask the user to promote the desired posts in the UI and then
+     retry `import_leads`.
      Source approval is the explicit confirmation for this bounded review
      batch; do not ask for a second yes/no gate between
      `select_promising_posts` and `import_leads`.
@@ -275,11 +286,14 @@ Shape:
 queue_cells({ tableId: workflowTableId, cellIds: reviewBatchEnrichCellIds })
 wait_for_campaign_table_ready({ tableId: workflowTableId })
 get_rows_minimal({ tableId: workflowTableId })
-wait_for_rubric_results({ tableId: workflowTableId, targetCount: cohortSize, includeRows: false })
+wait_for_rubric_results({ tableId: workflowTableId, targetCount: cohortSize, minPassedCount: minProjectedPass, includeRows: false })
 passInSample = count of first sampleSize review-batch rows with passesRubric === true
 projectedPass = round(passInSample / sampleSize * importLimit)
 
-if wait_for_rubric_results.ready === false and reason === "timeout":
+if wait_for_rubric_results.ready === true and partial === true and reason === "min_passed_count_met_with_resolved_failures":
+    use the partial passRate/stats as a valid sample diagnostic
+    continue if projectedPass >= minProjectedPass and generated messages are present for the passing rows
+else if wait_for_rubric_results.ready === false and reason === "timeout":
     use the partial passRate/stats as the sample diagnostic
     if active processing is visible:
         wait one more time at most

package/skills/providers/signal-discovery.md

@@ -377,6 +377,14 @@ selections, headlineICPCriteria })`, then call
 targetLeadCount })` for the approved review batch without asking for another
 yes/no gate. Do not import more than the approved bounded batch.
 
+The promotion/select step is required campaign state. Use post IDs from the
+current campaign-scoped search result, not stale IDs copied from a source review
+or scout summary. If `select_promising_posts` returns `selectedCount: 0`, stop:
+do not call `import_leads`, do not switch providers automatically, and do not
+claim the source failed. Re-run a narrow campaign-scoped search to recover
+current post rows, or have the user promote the posts in the UI and then retry
+the bounded import.
+
 ```json
 import_leads({
   "campaignOfferId": "cmp_xxx",