Merge pull request #5 from amplemarket/fixInconsistencies

Fix inconsistencies

Author: miguelaeh

skills/account-expansion-mapper/SKILL.md

@@ -1,14 +1,16 @@
 ---
 name: ae-pipeline-review
 description: >
-  Help AEs get a status check on their accounts with open deals — where things stand, where to focus, and what next steps to take.
+  Help AEs get a status check on their accounts with open deals -- where things stand, where to focus, and what next steps to take.
 metadata:
   author: amplemarket
-  version: "1.0.1"
+  version: "1.0.2"
 compatibility: Requires Amplemarket MCP server
 ---
 
-Help AEs get a status check on their accounts with open deals — where things stand, where to focus, and what next steps to take.
+# ae-pipeline-review
+
+Help AEs get a status check on their accounts with open deals -- where things stand, where to focus, and what next steps to take.
 
 ## Instructions
 
@@ -17,29 +19,29 @@ When an AE wants to review their accounts or deals, pull account data, engagemen
 ### Steps
 
 1. **Identify the AE's accounts.** Either:
-	- Use `mcp__claude_ai_Amplemarket__list_accounts` filtered by the AE's email to find accounts they own
-	- Filter for accounts with open opportunities if possible
-	- Or ask the AE for the specific accounts or domains they want reviewed
+    - Use `mcp__claude_ai_Amplemarket__list_accounts` filtered by the AE's email to find accounts they own
+    - Filter for accounts with open opportunities if possible
+    - Or ask the AE for the specific accounts or domains they want reviewed
 2. **Gather account data** by calling `mcp__claude_ai_Amplemarket__get_account` for each account. This gives engagement stats, AI-generated insights (interest level, current solution, competitors, pain points, champions, decision makers, gatekeepers, timelines), CRM data, and contacts.
 3. **Assess each account.** For each, evaluate:
-	- Deal momentum — is engagement progressing or stalled?
-	- Stakeholder coverage — are champions, decision makers, and economic buyers identified? Is the deal single-threaded?
-	- Risks — gone cold, competitor mentioned, champion left, missing key roles?
-	- Next steps — what should the AE do next?
+    - Deal momentum -- is engagement progressing or stalled?
+    - Stakeholder coverage -- are champions, decision makers, and economic buyers identified? Is the deal single-threaded?
+    - Risks -- gone cold, competitor mentioned, champion left, missing key roles?
+    - Next steps -- what should the AE do next?
 4. **Help with next steps when asked.** If the AE wants to act on a specific account:
-	- `mcp__claude_ai_Amplemarket__search_people` to find additional stakeholders for multi-threading
-	- `mcp__claude_ai_Amplemarket__get_contacts` to check status on specific contacts
-	- `mcp__claude_ai_Amplemarket__enrich_person` for deeper context on a key stakeholder
-5. **Present a well-formatted account review.** Prioritize by urgency — which accounts need action now vs progressing fine. For each, show current state, key people, and recommended next step.
+    - `mcp__claude_ai_Amplemarket__search_people` to find additional stakeholders for multi-threading
+    - `mcp__claude_ai_Amplemarket__get_contacts` to check status on specific contacts
+    - `mcp__claude_ai_Amplemarket__enrich_person` for deeper context on a key stakeholder
+5. **Present a well-formatted account review.** Prioritize by urgency -- which accounts need action now vs progressing fine. For each, show current state, key people, and recommended next step.
 
 ## Examples
 
 **User prompt:** "Give me a status on my deals"
+
 **Simple example output:** The agent pulls the AE's accounts, finds 3 needing immediate attention (one stalled in negotiation for 2 weeks, one single-threaded with only a junior contact, one where the champion seems to be going cold). 6 are progressing normally. Presented as a prioritized list with per-account summary and next steps.
 
 ## Troubleshooting
 
 | Problem | Solution |
-|---|---|
-| Too many accounts | Focus on accounts with open opportunities or recent engagement first. Offer to review the rest separately. |
-
+| --- | --- |
+| Too many accounts | Focus on accounts with open opportunities or recent engagement first. Offer to review the rest separately. |

skills/account-whitespace-analyzer/SKILL.md

@@ -4,10 +4,12 @@ description: >
   End-to-end workflow to search for prospects, preview results, refine criteria, and create a confirmed lead list in Amplemarket.
 metadata:
   author: amplemarket
-  version: "1.0.1"
+  version: "1.0.2"
 compatibility: Requires Amplemarket MCP server
 ---
 
+# build-targeted-lead-list
+
 # Build Targeted Lead List
 
 End-to-end workflow to search for prospects, preview results, refine criteria, and create a confirmed lead list in Amplemarket.
@@ -19,48 +21,52 @@ Guide the user through building a lead list from scratch. This is a multi-step,
 ### Steps
 
 1. **Gather search criteria** from the user. Ask for any of:
-	- Target titles and seniority levels
-	- Target departments or job functions
-	- Person locations
-	- Company names, domains, industries, sizes, or types
-	- Any other qualifying criteria
-	If the user gives a broad description, parse it into structured filters (see the prospect-icp-search skill for mapping guidance).
+    - Target titles and seniority levels
+    - Target departments or job functions
+    - Person locations
+    - Company names, domains, industries, sizes, or types
+    - Any other qualifying criteria
+    
+    If the user gives a broad description, parse it into structured filters (see the prospect-icp-search skill for mapping guidance).
+    
 2. **Resolve enum values** by calling:
-	- `mcp__claude_ai_Amplemarket__get_industries` to match industry terms to valid API values.
-	- `mcp__claude_ai_Amplemarket__get_job_functions` to match job function terms to valid API values.
+    - `mcp__claude_ai_Amplemarket__get_industries` to match industry terms to valid API values.
+    - `mcp__claude_ai_Amplemarket__get_job_functions` to match job function terms to valid API values.
 3. **Execute the search** by calling `mcp__claude_ai_Amplemarket__search_people` with the mapped parameters. Set `full_output` to `true` and `page_size` to 20 for the preview.
 4. **Present a preview** of the results to the user:
-	- Show the first 10-20 results in a table (Name, Title, Company, Location).
-	- Report the total number of matching prospects.
-	- Ask: "Does this look right? Would you like to refine the search or proceed with creating the list?"
+    - Show the first 10-20 results in a table (Name, Title, Company, Location).
+    - Report the total number of matching prospects.
+    - Ask: "Does this look right? Would you like to refine the search or proceed with creating the list?"
 5. **Refine if needed.** If the user wants changes:
-	- Adjust filters based on feedback.
-	- Re-run `mcp__claude_ai_Amplemarket__search_people` with updated parameters.
-	- Show the updated preview.
-	- Repeat until the user approves.
+    - Adjust filters based on feedback.
+    - Re-run `mcp__claude_ai_Amplemarket__search_people` with updated parameters.
+    - Show the updated preview.
+    - Repeat until the user approves.
 6. **Collect all results** once approved. If total results exceed one page, paginate through all pages using the `page` parameter to gather all LinkedIn URLs (up to the 10,000 lead limit).
 7. **Ask for list configuration:**
-	- **List name** (suggest a descriptive default like "VP Sales - Fintech - NYC - Mar 2026")
-	- **List type** - determine based on available data:
-		- `"linkedin"` if leads have LinkedIn URLs
-		- `"email"` if leads have email addresses
-		- `"titles_and_company"` if leads have title + company domain
-	- **Enrichment options:** Ask if they want to enable (note: each option consumes additional Amplemarket credits per lead):
-		- Email enrichment (`enrich`: true)
-		- Email validation (`validate_email`: true)
-		- Phone number reveal (`reveal_phone_numbers`: true)
-	- Inform the user of the credit implications before enabling enrichment: "Enabling enrichment will use credits for each lead. For a list of [N] leads, that's [N] enrichment credits. Want to proceed?"
+    - **List name** (suggest a descriptive default like "VP Sales - Fintech - NYC - Mar 2026")
+    - **List type** - determine based on available data:
+        - `"linkedin"` if leads have LinkedIn URLs
+        - `"email"` if leads have email addresses
+        - `"titles_and_company"` if leads have title + company domain
+    - **Enrichment options:** Ask if they want to enable (note: each option consumes additional Amplemarket credits per lead):
+        - Email enrichment (`enrich`: true)
+        - Email validation (`validate_email`: true)
+        - Phone number reveal (`reveal_phone_numbers`: true)
+    - Inform the user of the credit implications before enabling enrichment: "Enabling enrichment will use credits for each lead. For a list of [N] leads, that's [N] enrichment credits. Want to proceed?"
 8. **Create the lead list** by calling `mcp__claude_ai_Amplemarket__create_lead_list` with:
-	- `type`: determined list type
-	- `name`: chosen name
-	- `leads`: array of lead objects (with `linkedin_url`, `email`, or `title` + `company_domain` depending on type)
-	- `options`: enrichment settings
-	If total leads exceed the batch that fits in a single create call, use `mcp__claude_ai_Amplemarket__add_leads_to_lead_list` to add remaining leads to the created list using the returned `lead_list_id`.
+    - `type`: determined list type
+    - `name`: chosen name
+    - `leads`: array of lead objects (with `linkedin_url`, `email`, or `title` + `company_domain` depending on type)
+    - `options`: enrichment settings
+    
+    If total leads exceed the batch that fits in a single create call, use `mcp__claude_ai_Amplemarket__add_leads_to_lead_list` to add remaining leads to the created list using the returned `lead_list_id`.
+    
 9. **Confirm creation** with the user:
-	- Lead list ID
-	- Total leads added
-	- Enrichment settings enabled
-	- Estimated processing time
+    - Lead list ID
+    - Total leads added
+    - Enrichment settings enabled
+    - Estimated processing time
 
 ### Important Notes
 
@@ -74,6 +80,7 @@ Guide the user through building a lead list from scratch. This is a multi-step,
 ### Example 1: Full Workflow from Search to List
 
 **User prompt:** "Build me a lead list of engineering managers at healthcare companies in the US with 500-5000 employees"
+
 **What the skill does:**
 1. Calls `mcp__claude_ai_Amplemarket__get_industries` to resolve healthcare industry values.
 2. Calls `mcp__claude_ai_Amplemarket__search_people` with:
@@ -89,25 +96,29 @@ Guide the user through building a lead list from scratch. This is a multi-step,
 5. Calls `mcp__claude_ai_Amplemarket__create_lead_list` with:
 - `type`: "linkedin"
 - `name`: "Eng Managers - Healthcare US - 500-5K - Mar 2026"
-- `leads`: [{"linkedin_url": "…"}, …]
+- `leads`: [{"linkedin_url": "..."}, ...]
 - `options`: {"enrich": true, "validate_email": true}
+
 **Example confirmation:**
+
 Lead list created successfully.
 
 | Detail | Value |
-|---|---|
+| --- | --- |
 | List ID | lst_abc123def456 |
 | List Name | Eng Managers - Healthcare US - 500-5K - Mar 2026 |
 | Total Leads | 312 |
 | Type | LinkedIn |
 | Enrichment | Enabled |
 | Email Validation | Enabled |
 | Status | Processing |
+
 Your list is now being enriched. You can check its status anytime.
 
 ### Example 2: Quick List from Existing URLs
 
 **User prompt:** "Create a lead list called 'Q1 Targets' from these LinkedIn URLs: linkedin.com/in/user1, linkedin.com/in/user2, linkedin.com/in/user3"
+
 **What the skill does:**
 1. Skips search since the user has provided leads directly.
 2. Calls `mcp__claude_ai_Amplemarket__create_lead_list` with:
@@ -119,9 +130,10 @@ Your list is now being enriched. You can check its status anytime.
 ### Example 3: Refined Search Workflow
 
 **User prompt:** "Put together a list of senior marketing people at Series B+ SaaS companies in Europe"
+
 **What the skill does:**
 1. Resolves industries via `mcp__claude_ai_Amplemarket__get_industries`.
-2. Runs initial search with `person_seniorities`: ["Senior", "Manager", "Director", "VP", "Head"], `person_departments`: ["Marketing"], `person_locations`: ["United Kingdom", "Germany", "France", "Netherlands", "Sweden", …], matched SaaS industries.
+2. Runs initial search with `person_seniorities`: ["Senior", "Manager", "Director", "VP", "Head"], `person_departments`: ["Marketing"], `person_locations`: ["United Kingdom", "Germany", "France", "Netherlands", "Sweden", ...], matched SaaS industries.
 3. Shows preview. User says "too many junior people, only Directors and above."
 4. Re-runs with `person_seniorities`: ["Director", "VP", "Head", "C-Suite"].
 5. Shows updated preview. User approves.
@@ -130,12 +142,11 @@ Your list is now being enriched. You can check its status anytime.
 ## Troubleshooting
 
 | Problem | Solution |
-|---|---|
+| --- | --- |
 | Too many results to fit in one list | Paginate through results and use `mcp__claude_ai_Amplemarket__add_leads_to_lead_list` for batches after the first create call. |
 | List creation fails | Verify that leads array format matches the list type. LinkedIn lists need `linkedin_url`, email lists need `email`, titles_and_company lists need `title` + `company_domain`. |
 | Search returns irrelevant results | Tighten filters progressively: 1) Add `person_seniorities` to exclude junior titles. 2) Add `person_departments` to restrict to relevant functions. 3) Add `company_sizes` if results include wrong-size companies. Show the user a preview after each refinement so they can confirm the direction. |
 | User wants to add to existing list | Use `mcp__claude_ai_Amplemarket__list_lead_lists` to find the existing list, then `mcp__claude_ai_Amplemarket__add_leads_to_lead_list` with its ID. |
 | Enrichment not working | Check that the `options` object is correctly structured with boolean values for `enrich`, `validate_email`, and `reveal_phone_numbers`. |
 | `search_companies` returns 0 results for a known company | Fallback chain: 1) Try searching by company domain instead of name. 2) Try `enrich_company` with the domain directly, as some companies are enrichable but not searchable by name. 3) Try parent company domain or alternate domains. |
-| Search returns results but list creation fails | Verify that the leads array format matches the list type. LinkedIn lists need `linkedin_url`, email lists need `email`. |
-
+| Search returns results but list creation fails | Verify that the leads array format matches the list type. LinkedIn lists need `linkedin_url`, email lists need `email`. |