Research

Step 1 — Classify Targets

From "$ARGUMENTS", determine entity type and extract identifiers:

Business targets — when input looks like company names, domains, or URLs:

• Extract name (company name) and/or domain (domain or full URL stripped to domain)
• Build a JSON array: [{"name": "Salesforce", "domain": "salesforce.com"}, ...]

Prospect targets — when input looks like people (name + company, email, LinkedIn URL, phone):

• Extract full_name, company_name, email, phone_number, linkedin
• Build a JSON array: [{"full_name": "John Smith", "company_name": "Acme"}, ...]

CSV/JSON file input — when the argument is a file path, pass it directly to --from-file. Ask the user for the column mapping if it isn't obvious.

If the entity type is ambiguous, ask the user one clarifying question before continuing.

Step 2 — Match to Explorium IDs

Write the identifiers to a temp file and run match:

MATCH_OUT=$(mktemp /tmp/vibep-match.XXXXXX.json)

# For businesses (inline JSON)
node "$CLI" businesses match \
  --businesses '[{"name":"Salesforce","domain":"salesforce.com"}]' \
  --call-reasoning "$QUERY" > "$MATCH_OUT"

# For businesses (from file)
node "$CLI" businesses match \
  --from-file leads.csv \
  --column-map '{"Company":"name","Website":"domain"}' \
  --to-file "$MATCH_OUT" \
  --call-reasoning "$QUERY"

# For prospects (inline JSON)
node "$CLI" prospects match \
  --prospects '[{"full_name":"John Smith","company_name":"Acme","email":"[email protected]"}]' \
  --call-reasoning "$QUERY" > "$MATCH_OUT"

# For prospects (from file)
node "$CLI" prospects match \
  --from-file leads.csv \
  --column-map '{"Name":"full_name","Company":"company_name","Email":"email","LinkedIn":"linkedin"}' \
  --to-file "$MATCH_OUT" \
  --call-reasoning "$QUERY"

Inspect match results and surface any unmatched rows to the user now. Only proceed with matched rows (those containing business_id or prospect_id).

Step 3 — Enrich

Run enrich on the matched output, selecting enrichments appropriate to the entity type:

ENRICH_OUT=$(mktemp /tmp/vibep-enrich.XXXXXX.json)

# Businesses — firmographics + funding + tech stack + intent signals
node "$CLI" businesses enrich \
  --input "@$MATCH_OUT" \
  --enrichments firmographics,technographics,funding_and_acquisition,website_traffic,bombora_intent \
  --call-reasoning "$QUERY" > "$ENRICH_OUT"

# Prospects — contact info + social profiles
node "$CLI" prospects enrich \
  --input "@$MATCH_OUT" \
  --enrichments contacts_information,profiles \
  --call-reasoning "$QUERY" > "$ENRICH_OUT"

If the matched set contains both businesses and prospects, run enrich for each entity type separately and merge for presentation.

Step 4 — Fetch Trigger Events

Fetch recent events for each matched entity (last 90 days by default):

EVENTS_OUT=$(mktemp /tmp/vibep-events.XXXXXX.json)
SINCE=$(date -v-90d +%Y-%m-%d 2>/dev/null || date -d '90 days ago' +%Y-%m-%d)

# Businesses
node "$CLI" businesses events \
  --input "@$MATCH_OUT" \
  --event-types new_funding_round,new_partnership,new_product,hiring_in_engineering_department,hiring_in_sales_department,increase_in_all_departments,merger_and_acquisitions \
  --since "$SINCE" \
  --call-reasoning "$QUERY" > "$EVENTS_OUT"

# Prospects
node "$CLI" prospects events \
  --input "@$MATCH_OUT" \
  --event-types prospect_changed_role,prospect_changed_company,prospect_job_start_anniversary \
  --since "$SINCE" \
  --call-reasoning "$QUERY" > "$EVENTS_OUT"

Step 5 — Present the Dossier

For each successfully matched target, present one card:

[Company Name / Full Name] · [Match confidence if available]

[Title] · [Company] · [Industry] · [Employee range] employees

Recent trigger events (last 90 days):

For unmatched rows, show a brief "No Explorium record found" note.

Step 6 — Offer Next Actions

Ask the user which action to take:

1. Generate a list — Pivot to /vibeprospecting-plugin:generate-list using this research as ICP context
2. Expand research — Run deeper enrichments (e.g. workforce_trends, competitive_landscape, linkedin_posts)
3. Export to CSV — Re-run enrich with --to-file enriched.csv to produce a downloadable file
4. Research more targets — Add more names/domains/emails to the current session