GTM Account Research

Deep-dive a target company and produce a structured account intelligence brief covering tech stack, buying signals, key personas, and a recommended engagemen...

Security Scan

The skill is an instruction-only web-research template for producing GTM account briefs; its requested capabilities and absence of credentials/install steps are consistent with its stated purpose.

Purpose & Capability

Name and description match the SKILL.md steps: live web research (BuiltWith, job postings, blogs), identifying personas and signals, and producing a structured brief. The skill does not request unrelated binaries, credentials, or config paths — there is no disproportionate access requested for the stated task.

Instruction Scope

Instructions are scoped to public web research and grounding every signal in a traceable source. They instruct the agent to find people and LinkedIn URLs, date signals, and flag unverified items. This is in-scope but raises operational/privacy considerations: scraping or bulk collection of LinkedIn or other sites may run into terms-of-service, rate-limiting, or require authentication; outputs will include personal data (names + LinkedIn) which may have compliance implications.

Install Mechanism

No install spec and no code files — instruction-only. This is the lowest-risk install mechanism and matches the skill's purpose.

Credentials

The skill requests no environment variables, credentials, or config paths. That is proportionate: public web signals typically do not require secrets. Note: some desired signals (e.g., private job posting details or LinkedIn 'view' content) might be inaccessible without authenticated access, but the skill does not request those credentials.

Persistence & Privilege

always:false and default autonomous invocation are set. The skill does not request permanent presence or change other skills' configs. Autonomous invocation is normal; consider governance if you do not want the agent to autonomously perform live web lookups that may collect PII.

Assessment

This skill is coherent and instruction-only: it will perform live web research and produce briefs citing sources. Before installing, confirm the agent platform grants safe, rate-limited web search access and that you accept the agent collecting public personal data (names + LinkedIn URLs). Be aware of site terms-of-service (LinkedIn, job boards) and privacy/regulatory requirements (GDPR, CCPA) when storing or using identified personas. If you want to limit risk, keep autonomous invocation off or require human review, and add guardrails to avoid scraping behind logins or bulk-harvesting personal data. If you expect proprietary signals that require authenticated APIs, provide credentials explicitly and only if you trust the skill — currently it requests none.

Produce an Account Intelligence Brief for a named company. Ground every signal in a specific, traceable source.

Steps

1. Snapshot — company name, HQ, size, business model, recent news (last 90 days)
2. Tech stack — identify tools in the categories most relevant to your product (e.g. data warehouse, CRM, marketing automation, paid channels). Use BuiltWith, job postings, and engineering blogs as sources.
3. Buying signals — job postings, conference attendance, funding events, published strategy content. Date every signal.
4. Displacement triggers — are they actively migrating off a competitor? Hiring for a role that signals a gap you fill? Flag ⚡ if a strong trigger is present.
5. Personas — identify Economic Buyer (budget authority), Technical Champion (evaluates), and End User (daily use). Name + LinkedIn URL when findable.
6. Engagement angle — one specific recommendation: what pain, which product, what channel, who reaches out first, and the opening hook.

Output

# Account Brief: [Company] | [Date]

Snapshot: [3 sentences — what they do, size, recent context]
ICP Segment: [your segment label]
Stack: [Category: tool — or "Unknown" for each category]
Signals: [Bulleted — signal + source + date]
Displacement: [⚡ Yes — detail / None detected]
Personas: [Name · Title · LinkedIn · Pain point — one row per person]
Gaps: [What's missing that would sharpen the approach]

Engagement Angle:
[Specific: pain → product → channel → who reaches out → opening hook]

Keep the brief scannable. Flag any detail as [Unverified] if not directly observed.