菜单
Researches a company, person, buying committee, signal, domain, stack, financial context, or competitor mention. Produces a focused answer with sources and confidence notes.
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
Use this skill when the user wants to submit, contribute, or publish a GTM skill to the GTM Skills library (github.com/swan-gtm/gtm-skills, gtmskills.com), or update one of their existing skills or their creator profile there.
Use this skill when [the concrete trigger — a signal fires, a moment in the deal, a recurring cadence]. Produces [the concrete output]. [One or two more sentences dense with the phrases someone would actually say when they need this.]
Use this skill when a deal closes won and you want to turn the win into pipeline immediately. It extracts the win profile, finds lookalike companies not yet in your CRM, researches the right persona contact at each, drafts outreach from the deal owner using an anonymized win reference, and queues a single batch review task — nothing sends without human approval.
Use this skill when scoring a paying customer (closed-won account) for expansion. It evaluates product usage depth, persona quality, and expansion headroom to produce an expansion tier with short actionable reasoning. Designed as a sub-skill of your lead-scoring methodology — the parent runs the ACV assessment first and this returns a tier to it.
Use this skill when a profile-view webhook (e.g. LeadShark) fires for a new view on a team member's LinkedIn profile. It resolves the viewer's identity and company, applies dedup, internal, relationship, and ICP gates, and routes qualified viewers into the MQL pipeline — treating a profile view exactly like a website visit. The CRM is never polluted: only fully resolved, ICP-matched viewers reach a CRM write.
Use this skill when a tracked influencer publishes a LinkedIn post focused on your product (sponsored, review, or organic mention) and you want to capture the engagers as leads. It verifies the post is genuinely about you, scrapes the full engager list, filters to ICP-fit companies and buyer personas, runs lead scoring, and posts per-lead alerts plus a reconciled digest. Detection and scoring only — no outreach.
| name | research |
| title | Research |
| description | Researches a company, person, buying committee, signal, domain, stack, financial context, or competitor mention. Produces a focused answer with sources and confidence notes. |
| category | Research |
Research is target-driven and doesn't require org config to function — there's no state-hint paragraph and no Setup.md. The agent just routes to the right sub-page and runs.
Route based on what the user is asking about. Load the sub-page via swan-get-skill.
| Ask shape | Sub-page |
|---|---|
| Generic company / one company / "what does X do" / "/research acme" | Company |
| Named person — "/research tom from acme", "tell me about Jane Doe" | Person |
| Mapping the people inside one account — buying committee, who to thread to | AccountTeam |
| How does the target sell / their GTM / pricing posture | GTMmotion |
| What's recent / news / signals / last-30-days | News |
| Domain-level / web traffic / SEO / digital footprint | Domain |
| Financial health / funding / revenue / runway / ownership | Financial |
| Tech stack only — what tools do they run | TechStack |
| Where the target talks about competitors / displacement signals | CompetitorMentions |
If the ask is ambiguous (e.g. just a company name with no further context), default to Company. If the ask explicitly chains multiple facets ("research acme — company plus the buying committee"), it's fine to load two sub-pages — but always pick a primary.
Most asks want a one-paragraph answer; some want a dossier. Read the user's phrasing:
Default to brief. Scale up only when the user signals depth, or when the situation (board prep, founder intro, executive outreach) demands it. Over-researching when a sentence would do is the most common failure mode.
The sub-page owns the procedure for its facet. Follow it.
Across all sub-pages, the same cheap-before-expensive pattern applies:
swan-search-companies, swan-get-memory, prior briefs)swan-fetch-scraped-url, swan-fetch-businesses, swan-website-traffic, swan-fetch-business-events) before paid enrichmentswan-enrich-company, swan-enrich-contact) only when the cheap signals left a real gapDefault to a structured short summary, not raw scraped data. Every sub-page ends with a composed block — follow its shape. Always:
swan-website-traffic", "from LinkedIn post 2026-05-03", "press").swan-update-company when the research produced a durable fact worth reusing (financial state, tech stack, motion type) so the next play inherits it.What gets overlooked:
swan-fetch-scraped-url, swan-fetch-businesses, swan-website-traffic, swan-fetch-business-events) before chaining paid enrichments.swan-enrich-company or swan-enrich-contact calls without first checking what Swan and the CRM already have.files/tool-outputs/<toolName>_<callId>.json in swan-execute-code and summarize from there.Shipped today:
Company — general company research (the default landing for "/research acme")Person — full dossier on a named personAccountTeam — buying-committee mapping inside one accountGTMmotion — how the target sells todayNews — last-30-days signal scanDomain — digital footprint, traffic, SEO postureFinancial — funding history, ownership, runway, pressureTechStack — detected tools by categoryCompetitorMentions — public mentions of a named competitorAccreted over time as use cases sharpen:
AcquisitionTargets — research for M&A or partnership shortlistsCustomerHealth — research framing for an existing customer (expansion / churn risk)IndustryDeepDive — a whole vertical, not one accountInvestor — partner / fund researchBoard — board prep on a director or chair