outreach-personalization-research — independently scanned and version-tracked by SaferSkills.
SaferSkills independently audited outreach-personalization-research (Agent Skill) and scored it 100/100 (green). The audit ran 55 deterministic rules across Security, Supply Chain, Maintenance, Transparency, and Community; it found 0 high-severity and 0 lower-severity findings. The full rule-by-rule trace and per-finding evidence are below. Free, methodology-open.
Findings & checks · 0 flagged
Every scanned point with the score it earned and what moved between them.
First recorded scan — no prior version to compare against.
The primary manifest — the file an agent reads to learn what this artifact does.
Research a prospect deeply and generate actionable personalization angles with draft opening lines for cold outreach.
When a user wants to personalize outreach to a specific prospect, gather rich context and synthesize it into concrete angles and openers.
If the user provides minimal info, ask for at least a name and company or a LinkedIn URL.
mcp__claude_ai_Amplemarket__enrich_person with all available identifiers. Set reveal_email to true for contact details. Credit note: Enabling reveal_email and reveal_phone_numbers consumes additional Amplemarket credits per reveal. Key data points to extract:mcp__claude_ai_Amplemarket__enrich_company with the company domain or linkedin_url. Key data points to extract:mcp__claude_ai_Amplemarket__search_people with company_domains: [prospect's company domain], person_seniorities matching the prospect's level, page_size: 5 to understand the team structure around them.Prospect Summary (one-paragraph context)
Personalization Angles (3-5 angles, each with):
Recommended Best Angle with reasoning
Contact Details for executing the outreach
User prompt: "Help me personalize outreach to linkedin.com/in/sarah-chen-vp-marketing"
What the skill does:
mcp__claude_ai_Amplemarket__enrich_person with linkedin_url: "https://linkedin.com/in/sarah-chen-vp-marketing", reveal_email: true.mcp__claude_ai_Amplemarket__enrich_company with domain extracted from enrichment.mcp__claude_ai_Amplemarket__search_people with the company domain and seniority filter to understand team structure.Example output:
PERSONALIZATION BRIEF: Sarah Chen
Prospect Summary Sarah Chen is the VP of Marketing at CloudMetrics (cloudmetrics.io), a Series B cloud analytics company with 180 employees based in Austin, TX. She joined 8 months ago from HubSpot where she was a Senior Director of Demand Gen. She holds an MBA from Wharton. CloudMetrics recently raised a $45M Series B and appears to be scaling their go-to-market team.
Personalization Angles
1. New Role, New Strategy (Career Trajectory) - Strong Data: Sarah joined CloudMetrics 8 months ago as VP Marketing, coming from HubSpot. Opening line: "Sarah, transitioning from a marketing powerhouse like HubSpot to lead marketing at a Series B company is a big shift. I imagine you are rebuilding the demand gen playbook from scratch rather than inheriting one."
2. Post-Funding Growth (Company News) - Strong Data: CloudMetrics raised $45M Series B; headcount grew from 120 to 180 in 6 months. Opening line: "With CloudMetrics' recent Series B and the team nearly doubling, I'd guess the pressure to scale pipeline in lockstep with headcount growth is very real right now."
3. Tech Stack Gap (Technology) - Medium Data: CloudMetrics uses Salesforce, Marketo, and Google Analytics but no ABM or intent data platform detected. Opening line: "I noticed CloudMetrics is running Marketo and Salesforce but doesn't seem to have an intent data layer yet. Curious if that's on your roadmap as you scale outbound."
4. Team Building (Team Structure) - Medium Data: Search shows 3 marketing hires in the last 3 months, all junior/mid-level. Opening line: "Looks like you have been building out the marketing team quickly at CloudMetrics. When you are scaling a team that fast, having the right tooling in place before the team outgrows it can save a lot of rework."
5. Wharton Connection (Shared Context) - Soft Data: Sarah holds an MBA from Wharton. Opening line: "Fellow Wharton alum here. Saw you are leading marketing at CloudMetrics and wanted to reach out." (Only use if sender is also a Wharton alum.)
Recommended Best Angle: Angle 1 (New Role) or Angle 2 (Post-Funding Growth). Both are timely, specific, and connect to real pain points. Combine them for maximum impact.
Contact Details
User prompt: "Find personalization angles for James Park, CTO at Notion"
What the skill does:
mcp__claude_ai_Amplemarket__enrich_person with name: "James Park", company_name: "Notion", reveal_email: true.mcp__claude_ai_Amplemarket__enrich_company with domain: "notion.so".User prompt: "Write an opener for [email protected] - keep it casual and short"
What the skill does:
| Problem | Solution |
|---|---|
| Very sparse enrichment data | Fallback chain: 1) Try additional identifiers (LinkedIn URL, company domain) to improve match quality. 2) Focus angles on company-level data (industry, size, funding, hiring signals) when person data is thin. 3) Search for the prospect's peers at the same company to infer team structure and role context. 4) Be transparent: "Limited data on this person. Angles are based primarily on company signals." |
| No email found | Suggest LinkedIn InMail or connection request. Provide the draft opener adapted for LinkedIn format. |
| Angles feel generic | Ensure you are pulling specific data points. If enrichment is thin, recommend the user share any additional context they have about the prospect. |
| User wants more angles | Search for the prospect's content (posts, articles) or expand the company enrichment to find additional data points. |
| Wrong person enriched | Verify with the user by confirming title and company. Fallback: use company_domain instead of company_name for disambiguation, or ask for LinkedIn URL for exact match. |
| Person enrichment succeeds but company enrichment fails | Fallback chain: 1) Try enrich_company with domain instead of name. 2) Try LinkedIn company URL. 3) Use person-level data for role-based and career trajectory angles, and note: "Company data unavailable. Angles focus on the prospect's career and role." |
| Multiple people match the same name + company | Present all matches with titles and LinkedIn URLs. Ask the user to confirm before generating angles. Never generate personalization for the wrong person. |
Company exists but search_companies returns 0 results | Fallback chain: 1) Try enrich_company with domain directly. 2) Try the LinkedIn company URL. 3) Try parent company domain. Some companies are enrichable by domain but not searchable by name. |
~30 seconds. Free. No account. Every finding cites a rule and a line of evidence.