pre-call-research-brief — independently scanned and version-tracked by SaferSkills.
SaferSkills independently audited pre-call-research-brief (Agent Skill) and scored it 96/100 (green). The audit ran 55 deterministic rules across Security, Supply Chain, Maintenance, Transparency, and Community; it found 0 high-severity and 1 lower-severity findings. The full rule-by-rule trace and per-finding evidence are below. Free, methodology-open.
Findings & checks · 1 flagged
The text {match} tells the agent to skip the normal "ask the user first" gate. Used adversarially it removes the human-in-the-loop check before destructive or sensitive actions, turning a normally-gated agent into a fire-and-forget executor.
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.
Generate a comprehensive, structured research brief before any sales call by combining person enrichment, company intelligence, CRM history, web research, and competitive analysis into a single actionable document.
When a user wants to prepare for a sales call, gather deep context from every available source and synthesize it into an 8-section brief with talking points, landmines, and dynamic personalization fields.
If the user provides most of this context upfront, skip the questions that are already answered and only ask for what is missing. At minimum you need the person identifier and the call type to proceed.
mcp__claude_ai_Amplemarket__enrich_person with all available identifiers (linkedin_url, email, or name + company_name). Set reveal_email to true. Extract: current title, career history, education, skills, location, social profiles.WebSearch for the person's name + company to find recent activity: blog posts, podcast appearances, conference talks, LinkedIn posts, press quotes, and interviews.WebFetch on the 2-3 most relevant URLs from the search results to extract detailed content (quotes, topics discussed, positions taken).Key data to capture: career trajectory, communication style clues (formal vs. casual from their writing), topics they care about publicly, recent role changes.
mcp__claude_ai_Amplemarket__enrich_company with the company domain or linkedin_url. Extract: industry, employee count, revenue range, tech stack, headquarters, funding history, description.WebSearch for the company name + "news" to find: recent funding rounds, product launches, leadership changes, partnerships, acquisitions, layoffs, earnings reports.WebFetch on the 2-3 most relevant news articles to extract details.Key data to capture: growth trajectory, recent strategic moves, public challenges, market positioning.
mcp__claude_ai_HubSpot__get_user_details to verify the connection and check permissions.mcp__claude_ai_HubSpot__search_crm_objects with object_type: "contacts" and filter by the prospect's email to find existing contact records.mcp__claude_ai_HubSpot__search_crm_objects with object_type: "deals" and filter by the company name or associated contact to find deal history.mcp__claude_ai_HubSpot__get_crm_objects for each deal ID to retrieve full deal details: stage, amount, close date, associated notes, and activity timeline.If HubSpot is not connected or the call fails, skip this step entirely and omit the CRM History section from the brief. Do not prompt the user to connect HubSpot.
WebSearch for the company name + "uses" or "tech stack" or "tools" to find what software and competitors they currently use.enrich_company (step 3) with known competitors to your user's product.mcp__claude_ai_Amplemarket__search_people with:company_domains: [prospect's company domain]person_seniorities: matching and adjacent to the prospect's level (e.g., if prospect is VP, search for C-Suite, VP, and Director)page_size: 10full_output: trueMap the organizational structure around the prospect: who they report to, who reports to them, peers in adjacent functions. This helps identify multi-threading opportunities and internal champions or blockers.
Section 1: Executive Summary (1 paragraph) A concise overview of who you are meeting, what their company does, the call type, and the 2-3 most important things to know going in.
Section 2: Person Profile
Section 3: Company Snapshot
Section 4: CRM History (skip entirely if HubSpot is not connected)
Section 5: Talking Points (3-5 points tailored to the call type)
Section 6: Landmines to Avoid Sensitive topics to steer clear of: recent layoffs, bad press, failed products, executive departures, lawsuits, controversial company decisions. Sourced from web research. If no landmines are found, state that explicitly.
Section 7: Questions to Ask (3-5 insightful questions) Questions that demonstrate you did your research and open up valuable conversation threads. Tailored to the call type and informed by the research findings.
Section 8: Competitive Intel Known tools and platforms the company uses. Potential displacement targets. Integration angles. Competitive positioning notes relevant to the conversation.
| Field | Value | Source |
|---|---|---|
{{brief_person_name}} | [full name] | enrich_person |
{{brief_person_title}} | [current title] | enrich_person |
{{brief_company_name}} | [company name] | enrich_company |
{{brief_company_stage}} | [funding stage and amount] | enrich_company + WebSearch |
{{brief_company_size}} | [employee count] | enrich_company |
{{brief_recent_news}} | [most notable recent development] | WebSearch |
{{brief_career_highlight}} | [most relevant career detail] | enrich_person |
{{brief_podcast_mention}} | [podcast or talk appearance] | WebSearch |
{{brief_blog_post}} | [recent article or post] | WebSearch |
{{brief_tech_stack}} | [key tools and platforms] | enrich_company |
{{brief_competitor_used}} | [relevant competitor tool] | WebSearch |
{{brief_crm_history}} | [deal history summary] | HubSpot CRM |
{{brief_talking_point_1}} | [first talking point] | Analysis |
{{brief_talking_point_2}} | [second talking point] | Analysis |
{{brief_talking_point_3}} | [third talking point] | Analysis |
{{brief_question_1}} | [first research-backed question] | Analysis |
{{brief_question_2}} | [second research-backed question] | Analysis |
{{brief_landmine}} | [primary landmine to avoid] | WebSearch |
{{brief_followup_opener}} | [suggested follow-up email opener] | Analysis |
If a field has no data available, set its value to "Not available" and note the reason.
User prompt: "Prep me for a call with Sarah Chen, VP of Marketing at CloudMetrics. It's a discovery call. We sell an AI-powered sales engagement platform."
What the skill does:
mcp__claude_ai_Amplemarket__enrich_person with name: "Sarah Chen", company_name: "CloudMetrics", reveal_email: true.mcp__claude_ai_Amplemarket__enrich_company with domain: "cloudmetrics.io".WebSearch for "Sarah Chen CloudMetrics" and "CloudMetrics news 2026".WebFetch on relevant results (podcast appearance, funding announcement).mcp__claude_ai_HubSpot__get_user_details to check HubSpot connection.mcp__claude_ai_HubSpot__search_crm_objects for deals associated with CloudMetrics.mcp__claude_ai_Amplemarket__search_people at cloudmetrics.io for org context.WebSearch for "CloudMetrics tech stack" and "CloudMetrics tools".Example output:
PRE-CALL RESEARCH BRIEFCall: Discovery | Date: March 10, 2026 Prospect: Sarah Chen, VP of Marketing @ CloudMetrics Prepared for: [Your Name] | Your Solution: AI-powered sales engagement platform
1. Executive Summary
You are meeting with Sarah Chen, VP of Marketing at CloudMetrics, a Series B cloud analytics company based in Austin, TX with approximately 180 employees. Sarah joined CloudMetrics 8 months ago from HubSpot, where she spent 4 years as Senior Director of Demand Gen. CloudMetrics raised a $45M Series B in July 2025 and has been aggressively scaling their go-to-market team. This is a discovery call. Focus on understanding how their current outbound and demand gen processes are keeping up with their rapid growth, and whether their existing tooling can scale with them.
2. Person Profile
3. Company Snapshot
4. CRM History
5. Talking Points
6. Landmines to Avoid
7. Questions to Ask
8. Competitive Intel
Dynamic Fields Reference
| Field | Value | Source |
|---|---|---|
{{brief_person_name}} | Sarah Chen | enrich_person |
{{brief_person_title}} | VP of Marketing | enrich_person |
{{brief_company_name}} | CloudMetrics | enrich_company |
{{brief_company_stage}} | Series B, $45M raised (July 2025) | enrich_company + WebSearch |
{{brief_company_size}} | ~180 employees | enrich_company |
{{brief_recent_news}} | Launched self-serve analytics dashboard (Feb 2026) | WebSearch |
{{brief_career_highlight}} | Senior Director of Demand Gen at HubSpot (4 years) | enrich_person |
{{brief_podcast_mention}} | Guest on SaaS Metrics Podcast (Jan 2026), discussed PLG growth | WebSearch |
{{brief_blog_post}} | "Why Your Demand Gen Strategy Needs to Break Before It Scales" (LinkedIn, Feb 2026) | WebSearch |
{{brief_tech_stack}} | Salesforce, Marketo, Snowflake, Google Analytics, Intercom | enrich_company |
{{brief_competitor_used}} | Outreach (sales engagement - inherited, not fully bought in) | HubSpot CRM + WebSearch |
{{brief_crm_history}} | 2 previous deals, both closed-lost (Jun 2025, Nov 2025) | HubSpot CRM |
{{brief_talking_point_1}} | Inherited Outreach - she did not choose it, fresh eval opportunity | Analysis |
{{brief_talking_point_2}} | Post-Series B scaling pressure on go-to-market team | Analysis |
{{brief_talking_point_3}} | Self-serve launch requires new pipeline motion | Analysis |
{{brief_question_1}} | How has your "no enterprise playbook" philosophy shaped demand gen here? | Analysis |
{{brief_question_2}} | How are you balancing PLG and outbound workflows? | Analysis |
{{brief_landmine}} | Do not mention January 2026 engineering layoffs (unverified, sensitive) | WebSearch |
{{brief_followup_opener}} | Great speaking today, Sarah. Following up on our conversation about scaling demand gen at CloudMetrics | Analysis |
User prompt: "I have a demo with Alex Rivera, Head of Revenue Operations at FinFlow tomorrow. We're showing them our analytics dashboard."
What the skill does:
Key differences from a discovery brief:
User prompt: "Meeting prep for my renewal call with David Kim, CTO at NexaHealth. Their contract is up next month and I want to expand to their European team."
What the skill does:
Key differences from a discovery brief:
| Problem | Solution |
|---|---|
| HubSpot not connected | Graceful degradation: skip CRM History (Section 4) entirely. Do not show an error or empty section. Add a note at the end of the brief: "CRM history not available. HubSpot integration not detected. Connect HubSpot for deal history, engagement timelines, and notes from previous conversations." The remaining 7 sections should still provide a comprehensive brief. |
| Sparse person data from enrichment | Fallback chain: 1) Try alternative identifiers. If you used name + company, try LinkedIn URL or email instead. 2) Lean heavily on WebSearch for the person's public activity (LinkedIn posts, conference talks, articles, podcast appearances). 3) Use company-level data and role-based assumptions to fill gaps. 4) Be transparent in the brief: "Limited person data available. Profile sections are based primarily on web research and role-based analysis." |
| No recent news found for the company | State this explicitly in the Company Snapshot: "No significant news coverage found in the last 6 months." Shift focus to enrichment data (funding history, employee growth trends, tech stack changes). Check the company's own blog and press page via WebFetch as a secondary source. |
| Person has a common name (disambiguation) | If enrich_person returns results that do not match the expected company or title, do not proceed with bad data. Ask the user: "I found multiple matches for [name]. Can you confirm their LinkedIn URL or email for an exact match?" If the user cannot provide it, use company_domain as a filter and present the best match with a confidence note. |
| enrich_person returns the wrong person | Verify the returned profile against the user's description (company, title). If it does not match: 1) Re-try with LinkedIn URL if available. 2) Try with email address. 3) Try with company_domain instead of company_name. 4) If still wrong, inform the user and build the brief from web research and company enrichment only, noting: "Person enrichment returned an incorrect match. Person profile is based on web research only." |
| WebFetch blocked or returns errors | Some websites block automated fetching. Fallback: 1) Use the WebSearch snippet text, which often contains enough context. 2) Try fetching the Google Cache version. 3) Note in the brief which sources could not be fully accessed: "[Article could not be fully retrieved - summary based on search snippet]." |
| Rate limiting on batch enrichment | If you hit rate limits when enriching multiple people in the team context step (step 6), reduce the page_size to 5 and prioritize the prospect's direct reports and manager. Inform the user: "Rate limits reached. Team context is based on a smaller sample. I can research additional team members separately if needed." |
The following fields are generated with every brief. Use them in email templates, CRM updates, or follow-up sequences.
| Field | Example Value | Source |
|---|---|---|
{{brief_person_name}} | "Sarah Chen" | enrich_person |
{{brief_person_title}} | "VP of Marketing" | enrich_person |
{{brief_company_name}} | "CloudMetrics" | enrich_company |
{{brief_company_stage}} | "Series B, $45M raised" | enrich_company + WebSearch |
{{brief_company_size}} | "180 employees" | enrich_company |
{{brief_recent_news}} | "Launched new analytics dashboard in Feb 2026" | WebSearch |
{{brief_career_highlight}} | "Previously Senior Director at HubSpot for 4 years" | enrich_person |
{{brief_podcast_mention}} | "Guest on SaaS Metrics Podcast, discussed PLG growth" | WebSearch |
{{brief_blog_post}} | "Published article on demand gen scaling challenges" | WebSearch |
{{brief_tech_stack}} | "Salesforce, Marketo, Snowflake" | enrich_company |
{{brief_competitor_used}} | "Currently using Outreach for sales engagement" | WebSearch |
{{brief_crm_history}} | "2 previous deals, last closed-lost Q3 2025" | HubSpot CRM |
{{brief_talking_point_1}} | "Their recent product launch creates new pipeline needs" | Analysis |
{{brief_talking_point_2}} | "Post-Series B scaling pressure on go-to-market team" | Analysis |
{{brief_talking_point_3}} | "Team grew 50% in 6 months - tooling may be outpaced" | Analysis |
{{brief_question_1}} | "How has the team's workflow changed since the Series B?" | Analysis |
{{brief_question_2}} | "What's your biggest challenge scaling demand gen now?" | Analysis |
{{brief_landmine}} | "Avoid mentioning recent layoffs in engineering (Jan 2026)" | WebSearch |
{{brief_followup_opener}} | "Great speaking today. Following up on the analytics discussion" | Analysis |
~30 seconds. Free. No account. Every finding cites a rule and a line of evidence.