{"id":13538,"plugin_id":"plugin_asdk_app_6a8c88f806c08191ab0ae925be246c65","kind":"skill","collection_source":null,"comparison_source":null,"observed_at":"2026-09-30T23:07:40.729Z","digest":"2b084a4eb698777a00640219a9f9da1656a792be0943418dadc25abfd48d7440","against":null,"payload":{"description":"Discovery with Unify, building lists of companies or people from criteria - ICP filters, personas, lookalikes, hiring signals, technographics, local businesses, funding stage. Use when the user wants to find prospects, build a target-account or contact list, expand TAM, or ask \"who should I sell to\".","included_files":[],"name":"discovery","skill_md_contents":"---\nname: discovery\ndescription: Discovery with Unify, building lists of companies or people from criteria - ICP filters, personas, lookalikes, hiring signals, technographics, local businesses, funding stage. Use when the user wants to find prospects, build a target-account or contact list, expand TAM, or ask \"who should I sell to\".\n---\n\n# Discovery: building lists\n\nDiscovery briefs go through `run_agent` (see `agent-runs` for the loop). The Unify\nagent routes to sources itself; your job is a precise brief.\n\n## Brief anatomy\n\nState, in order:\n\n1. **Entity type**: companies, people, or companies-then-people (\"find 20 fintech\n   companies, then 2–3 engineering leaders at each\").\n2. **Hard filters** (exact, verifiable): industry, headcount range, revenue,\n   geography (defaults to US, always state it), funding stage, tech stack, hiring\n   status.\n3. **Semantic intent** (fuzzy, persona-level): \"developer-tools buyers\",\n   \"operators who own retention\", natural-language titles. Unify's semantic search\n   over its proprietary dataset handles these well; don't flatten them into rigid\n   title lists yourself.\n4. **Target count** and **deliverable**: \"return N results as a DataTable\".\n5. **Exclusions**: existing customers, competitors, records already in the CRM or\n   a List (\"exclude companies already in our Salesforce\").\n\n## What Unify is good at (helps you scope)\n\n- **Universal Data** (free, proprietary): identity, domain, LinkedIn, industry, geo,\n  headcount, revenue, titles, work history, education (for 1.1B+ people and 65M+\n  companies). Prefer briefs answerable here when budget matters.\n- **Paid vendor signals** (credits per record): lookalike companies, hiring/job\n  postings, technographics, ecommerce/store data, local businesses (Google\n  Maps/Yelp), funding and venture data, web traffic/SEO, social buying signals.\n- **Weak/absent**: normalized seniority levels, live ad spend detail; expect the\n  agent to approximate or ask.\n\n## Patterns\n\n- **Scout before scale**: for big lists, first ask for 5–10 results to validate\n  criteria with your user, then a follow-up run for the full list.\n- **Lookalikes**: \"find companies similar to acme.com, stripe.com\" is a first-class\n  ask.\n- **Existing-data first**: \"which of our CRM accounts match X\" is a discovery brief\n  too. Unify joins CRM with engagement/intent signals for free.\n- Results land in a DataTable; page it with `load_datatable` and present a readable\n  sample, not the whole dump.\n"},"changes":[],"summary":"First saved snapshot. No earlier version is available for comparison.","summary_kind":"deterministic","summary_metadata":{}}