nyne-search（奈恩搜索）

Nyne Search Skill

Search for people using natural language queries. Find professionals by role, company, location, industry, or any combination. Returns matching profiles with contact info, work history, education, and optional AI relevance scoring.

Important: This API is async. You POST to submit, get a request_id, then poll GET until status: "completed". Light searches complete in 3-40 seconds; premium searches take 30-300 seconds.

Agent Instructions

When presenting search results to the user, show all returned data for each person. Walk through:

1. Result count — total_stored, total_estimate, has_more, credits_charged
2. Each person — displayname, headline, bio, location, gender, estimated_age, total_experience_years, is_decision_maker
3. Contact info — best_business_email, best_personal_email, altemails, fullphone (if show_emails/show_phone_numbers were enabled)
4. Social profiles — LinkedIn URL, username, connections, followers
5. Work history — all organizations with title, dates, company details (industries, num_employees, funding, technologies)
6. Education — schools with degree, major, dates; note is_top_universities flag
7. Interests — work interests and certifications
8. Patents — title, date, reference, URL (if present)
9. Languages — spoken languages
10. Score — AI relevance score 0-1 (if profile_scoring was enabled)
11. Insights — AI-generated match reasoning (if insights were enabled)

If has_more is true, tell the user there are more results available and offer to paginate using the next_cursor.

If a field is missing from the response, it means no data was found — skip it silently.

Setup

Required environment variables:

• NYNE_API_KEY — your Nyne API key
• NYNE_API_SECRET — your Nyne API secret

Get credentials at https://api.nyne.ai.

export NYNE_API_KEY="your-api-key"
export NYNE_API_SECRET="your-api-secret"

Verify they're set:

echo "Key: ${NYNE_API_KEY:0:8}... Secret: ${NYNE_API_SECRET:0:6}..."

Important: JSON Handling

The API response can contain control characters in JSON string values that break jq. All examples use a nyne_parse helper that cleans and re-encodes JSON via python3. Define it once per session:

nyne_parse() {
  python3 -c "
import sys, json, re
raw = sys.stdin.read()
clean = re.sub(r'[\-\]+', ' ', raw)
data = json.loads(clean)
json.dump(data, sys.stdout)
"
}

Quick Start

Search for people by natural language query and poll until complete:

nyne_parse() {
  python3 -c "
import sys, json, re
raw = sys.stdin.read()
clean = re.sub(r'[\-\]+', ' ', raw)
data = json.loads(clean)
json.dump(data, sys.stdout)
"
}

# Submit search request
curl -s -X POST "https://api.nyne.ai/person/search" \
  -H "Content-Type: application/json" \
  -H "X-API-Key: $NYNE_API_KEY" \
  -H "X-API-Secret: $NYNE_API_SECRET" \
  -d '{"query": "Software engineers at Google in San Francisco", "limit": 10, "type": "premium", "show_emails": true}' | nyne_parse > /tmp/nyne_search.json

STATUS=$(jq -r '.data.status' /tmp/nyne_search.json)

if [ "$STATUS" = "completed" ]; then
  echo "Search completed immediately."
  jq '.data | {total_stored, total_estimate, has_more, credits_charged}' /tmp/nyne_search.json
  jq '.data.results[] | {displayname, headline, location}' /tmp/nyne_search.json
else
  REQUEST_ID=$(jq -r '.data.request_id' /tmp/nyne_search.json)
  echo "Request submitted: $REQUEST_ID (status: $STATUS)"

  # Poll until complete (checks every 5s, times out after 10 min)
  SECONDS_WAITED=0
  while [ $SECONDS_WAITED -lt 600 ]; do
    curl -s "https://api.nyne.ai/person/search?request_id=$REQUEST_ID" \
      -H "X-API-Key: $NYNE_API_KEY" \
      -H "X-API-Secret: $NYNE_API_SECRET" | nyne_parse > /tmp/nyne_search.json
    STATUS=$(jq -r '.data.status' /tmp/nyne_search.json)
    echo "Status: $STATUS ($SECONDS_WAITED seconds elapsed)"
    if [ "$STATUS" = "completed" ]; then
      jq '.data | {total_stored, total_estimate, has_more, credits_charged}' /tmp/nyne_search.json
      jq '.data.results[] | {displayname, headline, location}' /tmp/nyne_search.json
      break
    elif [ "$STATUS" = "failed" ]; then
      echo "Search failed."
      jq . /tmp/nyne_search.json
      break
    fi
    sleep 5
    SECONDS_WAITED=$((SECONDS_WAITED + 5))
  done

  if [ $SECONDS_WAITED -ge 600 ]; then
    echo "Timed out after 10 minutes. Resume polling with request_id: $REQUEST_ID"
  fi
fi

Submit Search (POST)

Endpoint: POST https://api.nyne.ai/person/search

Headers:

Content-Type: application/json
X-API-Key: $NYNE_API_KEY
X-API-Secret: $NYNE_API_SECRET

Query Parameters

Custom Filters

Pass as a custom_filters object to narrow results beyond the natural language query.

Array fields (pass as arrays of strings):

Numeric range fields (pass as integers):

Exact match fields:

Examples

Basic search:

curl -s -X POST "https://api.nyne.ai/person/search" \
  -H "Content-Type: application/json" \
  -H "X-API-Key: $NYNE_API_KEY" \
  -H "X-API-Secret: $NYNE_API_SECRET" \
  -d '{"query": "Software engineers at Google in San Francisco", "limit": 10}' | nyne_parse > /tmp/nyne_search.json

jq '{status: .data.status, request_id: .data.request_id}' /tmp/nyne_search.json

With emails and phone numbers:

curl -s -X POST "https://api.nyne.ai/person/search" \
  -H "Content-Type: application/json" \
  -H "X-API-Key: $NYNE_API_KEY" \
  -H "X-API-Secret: $NYNE_API_SECRET" \
  -d '{"query": "Marketing directors at tech startups in Austin", "limit": 25, "show_emails": true, "show_phone_numbers": true}' | nyne_parse > /tmp/nyne_search.json

With custom filters:

curl -s -X POST "https://api.nyne.ai/person/search" \
  -H "Content-Type: application/json" \
  -H "X-API-Key: $NYNE_API_KEY" \
  -H "X-API-Secret: $NYNE_API_SECRET" \
  -d '{
    "query": "Engineering leaders in fintech",
    "limit": 20,
    "type": "premium",
    "show_emails": true,
    "custom_filters": {
      "locations": ["New York", "San Francisco"],
      "titles": ["CTO", "VP Engineering", "Head of Engineering"],
      "min_total_experience_years": 10,
      "industries": ["Financial Services", "Technology"]
    }
  }' | nyne_parse > /tmp/nyne_search.json

Light search (fast, 1 credit):

curl -s -X POST "https://api.nyne.ai/person/search" \
  -H "Content-Type: application/json" \
  -H "X-API-Key: $NYNE_API_KEY" \
  -H "X-API-Secret: $NYNE_API_SECRET" \
  -d '{"query": "Data scientists at Amazon", "type": "light", "limit": 10}' | nyne_parse > /tmp/nyne_search.json

With AI scoring and insights:

curl -s -X POST "https://api.nyne.ai/person/search" \
  -H "Content-Type: application/json" \
  -H "X-API-Key: $NYNE_API_KEY" \
  -H "X-API-Secret: $NYNE_API_SECRET" \
  -d '{"query": "AI researchers in healthcare", "type": "premium", "profile_scoring": true, "insights": true, "limit": 10}' | nyne_parse > /tmp/nyne_search.json

With freshness priority:

curl -s -X POST "https://api.nyne.ai/person/search" \
  -H "Content-Type: application/json" \
  -H "X-API-Key: $NYNE_API_KEY" \
  -H "X-API-Secret: $NYNE_API_SECRET" \
  -d '{"query": "Product managers who recently changed jobs", "type": "premium", "high_freshness": true, "show_emails": true}' | nyne_parse > /tmp/nyne_search.json

Require contact info:

curl -s -X POST "https://api.nyne.ai/person/search" \
  -H "Content-Type: application/json" \
  -H "X-API-Key: $NYNE_API_KEY" \
  -H "X-API-Secret: $NYNE_API_SECRET" \
  -d '{"query": "Sales directors at SaaS companies", "require_emails": true, "show_emails": true, "limit": 50}' | nyne_parse > /tmp/nyne_search.json

Submit response:

{
  "success": true,
  "data": {
    "request_id": "abc12345_1737123456_1234",
    "status": "completed",
    "results": [...],
    "offset": 0,
    "limit": 10,
    "total_stored": 47,
    "total_estimate": 2500,
    "has_more": true,
    "next_cursor": "eyJvIjoxMCwiciI6ImFiYzEyMzQ1XzE3MzcxMjM0NTZfMTIzNCJ9",
    "from_cache": false,
    "credits_charged": 7
  },
  "timestamp": "2026-02-19T..."
}

Note: The POST may return status: "completed" immediately (especially for cached or light searches) or status: "processing" requiring polling. Always check the status.

Poll for Results (GET)

Endpoint: GET https://api.nyne.ai/person/search?request_id={id}

Headers: Same X-API-Key and X-API-Secret as above.

Check status once

curl -s "https://api.nyne.ai/person/search?request_id=$REQUEST_ID" \
  -H "X-API-Key: $NYNE_API_KEY" \
  -H "X-API-Secret: $NYNE_API_SECRET" | nyne_parse > /tmp/nyne_search.json

jq '{status: .data.status, total_stored: .data.total_stored}' /tmp/nyne_search.json

Full polling loop

SECONDS_WAITED=0
TIMEOUT=600  # 10 minutes

while [ $SECONDS_WAITED -lt $TIMEOUT ]; do
  curl -s "https://api.nyne.ai/person/search?request_id=$REQUEST_ID" \
    -H "X-API-Key: $NYNE_API_KEY" \
    -H "X-API-Secret: $NYNE_API_SECRET" | nyne_parse > /tmp/nyne_search.json
  STATUS=$(jq -r '.data.status' /tmp/nyne_search.json)

  echo "[$(date +%H:%M:%S)] Status: $STATUS ($SECONDS_WAITED s)"

  case "$STATUS" in
    completed)
      jq '.data | {total_stored, total_estimate, has_more, credits_charged}' /tmp/nyne_search.json
      jq '.data.results[] | {displayname, headline, location}' /tmp/nyne_search.json
      break
      ;;
    failed)
      echo "Search failed:"
      jq '.data' /tmp/nyne_search.json
      break
      ;;
    *)
      sleep 5
      SECONDS_WAITED=$((SECONDS_WAITED + 5))
      ;;
  esac
done

if [ "$STATUS" != "completed" ] && [ "$STATUS" != "failed" ]; then
  echo "Timed out. Resume polling with request_id: $REQUEST_ID"
fi

Pagination

After the initial search completes, paginate through results for free (cached results, no additional credits).

Cursor-based pagination (recommended)

Use the next_cursor from the previous response:

curl -s -X POST "https://api.nyne.ai/person/search" \
  -H "Content-Type: application/json" \
  -H "X-API-Key: $NYNE_API_KEY" \
  -H "X-API-Secret: $NYNE_API_SECRET" \
  -d '{"cursor": "eyJvIjoxMCwiciI6ImFiYzEyMzQ1XzE3MzcxMjM0NTZfMTIzNCJ9"}' | nyne_parse > /tmp/nyne_search.json

jq '.data | {total_stored, offset, limit, has_more}' /tmp/nyne_search.json
jq '.data.results[] | {displayname, headline, location}' /tmp/nyne_search.json

Offset-based pagination

Use request_id + offset from the original search:

curl -s -X POST "https://api.nyne.ai/person/search" \
  -H "Content-Type: application/json" \
  -H "X-API-Key: $NYNE_API_KEY" \
  -H "X-API-Secret: $NYNE_API_SECRET" \
  -d '{"request_id": "abc12345_1737123456_1234", "offset": 25, "limit": 25}' | nyne_parse > /tmp/nyne_search.json

jq '.data | {total_stored, offset, limit, has_more}' /tmp/nyne_search.json
jq '.data.results[] | {displayname, headline, location}' /tmp/nyne_search.json

Pagination notes:

• Paginating through cached results is free — no additional credits charged
• Maximum 1000 results per search (offset + limit ≤ 1000)
• has_more indicates whether more results are available
• next_cursor is only present when has_more is true
• Results are cached for 30 days

Response Structure

When status is completed:

{
  "success": true,
  "timestamp": "2026-02-19T...",
  "data": {
    "request_id": "abc12345_1737123456_1234",
    "status": "completed",
    "results": [
      {
        "displayname": "Jane Smith",
        "headline": "Senior Software Engineer at Google",
        "bio": "...",
        "location": "San Francisco, CA",
        "gender": "female",
        "estimated_age": 35,
        "total_experience_years": 12,
        "is_decision_maker": 1,
        "is_top_universities": true,
        "languages": ["English", "Spanish"],
        "interests": {
          "work": ["software engineering", "machine learning", "distributed systems"],
          "certification": ["AWS Solutions Architect"]
        },
        "patents": [],
        "best_business_email": "jane@google.com",
        "best_personal_email": "jane@gmail.com",
        "altemails": ["j.smith@gmail.com"],
        "fullphone": [{"fullphone": "+14155551234", "type": "mobile"}],
        "social_profiles": {
          "linkedin": {
            "url": "https://www.linkedin.com/in/janesmith",
            "username": "janesmith",
            "connections": 1243,
            "followers": 542,
            "photo_url": "..."
          }
        },
        "organizations": [
          {
            "name": "Google",
            "title": "Senior Software Engineer",
            "startDate": "3-2020",
            "location": "San Francisco",
            "company_website": "google.com",
            "industries": ["Technology"],
            "num_employees": "10001+"
          }
        ],
        "schools_attended": ["Stanford University"],
        "schools_info": [
          {
            "name": "Stanford University",
            "degree": "Master's",
            "title": "Computer Science",
            "startDate": "2012",
            "endDate": "2014"
          }
        ],
        "score": 0.92,
        "insights": {
          "match_reasoning": "Strong match based on..."
        }
      }
    ],
    "offset": 0,
    "limit": 10,
    "total_stored": 47,
    "total_estimate": 2500,
    "has_more": true,
    "next_cursor": "eyJvIjoxMCwi...",
    "from_cache": false,
    "credits_charged": 7
  }
}

Result Fields

Social Profiles Object

{
  "linkedin": {
    "url": "https://www.linkedin.com/in/...",
    "username": "...",
    "connections": 1243,
    "followers": 542,
    "photo_url": "..."
  }
}

Organization Object

{
  "name": "Windfall Bio",
  "title": "Vice President of Software Engineering",
  "startDate": "2024-08-01",
  "endDate": null,
  "location": "San Mateo, California, United States",
  "company_website": "http://www.windfall.bio",
  "company_linkedin_url": "https://www.linkedin.com/company/windfallbio",
  "company_domain": "windfall.bio",
  "industries": ["Manufacturing"],
  "num_employees": 37,
  "num_employees_range": "11-50",
  "is_startup": true,
  "is_b2b": true,
  "is_b2c": false,
  "is_saas": false,
  "founded_in": 2022,
  "latest_funding_round": "Series A",
  "latest_funding_amount": 28000000,
  "funding_total_usd": 37000000,
  "annual_revenue": 345000,
  "specialties": ["methane", "climate", "agtech"],
  "technologies": ["Google Analytics", "AWS Lambda", "Webflow"],
  "companyDesc": "...",
  "logo_url": "..."
}

Organizations include rich company intelligence: funding data, tech stack, employee counts, and B2B/B2C/SaaS flags. Not all fields are present for every organization.

Response Metadata Fields

Useful jq Filters

After polling completes, the clean response is at /tmp/nyne_search.json:

# Search metadata
jq '.data | {total_stored, total_estimate, has_more, credits_charged, from_cache}' /tmp/nyne_search.json

# List all people (name + headline + location)
jq '.data.results[] | {displayname, headline, location}' /tmp/nyne_search.json

# All emails
jq '.data.results[] | {displayname, best_business_email, best_personal_email, altemails}' /tmp/nyne_search.json

# Phone numbers
jq '.data.results[] | {displayname, fullphone}' /tmp/nyne_search.json

# LinkedIn profiles
jq '.data.results[] | {displayname, linkedin: .social_profiles.linkedin.url}' /tmp/nyne_search.json

# Work history per person
jq '.data.results[] | {displayname, orgs: [.organizations[] | {name, title}]}' /tmp/nyne_search.json

# Decision makers only
jq '[.data.results[] | select(.is_decision_maker == true)] | length' /tmp/nyne_search.json
jq '[.data.results[] | select(.is_decision_maker == true) | {displayname, headline}]' /tmp/nyne_search.json

# Sort by relevance score (if profile_scoring enabled)
jq '[.data.results | sort_by(-.score)[] | {displayname, headline, score}]' /tmp/nyne_search.json

# Experience range
jq '.data.results[] | {displayname, total_experience_years} | select(.total_experience_years >= 10)' /tmp/nyne_search.json

# Education
jq '.data.results[] | {displayname, schools: .schools_attended}' /tmp/nyne_search.json

# Interests and certifications
jq '.data.results[] | {displayname, work_interests: .interests.work, certifications: .interests.certification}' /tmp/nyne_search.json

# Patents
jq '.data.results[] | select(.patents | length > 0) | {displayname, patents: [.patents[] | {title, date}]}' /tmp/nyne_search.json

# Company intelligence (funding, tech stack)
jq '.data.results[] | {displayname, companies: [.organizations[] | {name, funding_total_usd, latest_funding_round, num_employees, is_startup}]}' /tmp/nyne_search.json

# Get next_cursor for pagination
jq -r '.data.next_cursor' /tmp/nyne_search.json

# Count results
jq '.data.results | length' /tmp/nyne_search.json

Error Handling

Search Tiers & When to Use Each

Fast Search — type: "light" (1 credit per result)

Basic search with fast results in 3-40 seconds. Best for quick lookups when you need a broad list of names and don't need deep profile data. AI features (insights, high_freshness) are NOT available for light type. Profile scoring is available.

{"query": "Software engineers at Google", "type": "light", "limit": 50}

Smart Search — type: "medium"

Balanced tier between speed and quality. Use when you want better matching than light but don't need the full depth of premium. Supports all AI features.

{"query": "Engineering managers in fintech", "type": "medium", "limit": 25}

Pro Search — type: "premium" (5 credits per result)

Comprehensive search taking 30-300 seconds. Best quality results with deepest profile matching. Required for maximum-quality insights and high_freshness results. Use when result quality matters more than speed.

{"query": "AI researchers at top universities", "type": "premium", "limit": 10, "insights": true, "profile_scoring": true}

AI-Powered Analysis Features

These features add intelligence on top of search results. Insights and High Freshness are NOT available for light type searches — use medium or premium.

Query Insights (insights: true, +1 credit per result)

AI-generated analysis for each result explaining why this person matches your search query. Returns structured reasoning about the match relevance — how their background, role, and experience align with what you're looking for. Invaluable for qualifying leads or evaluating candidates at scale. Use when you need to understand the "why" behind each match, not just the "who".

{"query": "AI researchers in healthcare", "type": "premium", "insights": true, "limit": 10}

Profile Scoring (profile_scoring: true, +1 credit)

AI relevance scoring that assigns each result a score from 0 to 1, where 1.0 is a perfect match. Use to rank and prioritize results by relevance. Works with all search tiers including light. Combine with insights for both a score and an explanation.

{"query": "Senior engineers at fintech startups", "profile_scoring": true, "limit": 25}

High Freshness (high_freshness: true, +2 credits)

Prioritizes profiles that have been recently updated — people who recently changed jobs, updated their LinkedIn, or have fresh activity. Use when recency matters: hiring recently active candidates, targeting people who just moved to a new role, or finding profiles with current contact info. Not available for light type.

{"query": "Product managers who recently changed jobs", "type": "premium", "high_freshness": true, "show_emails": true}

Combining AI Features

Stack insights + scoring + freshness for maximum intelligence:

{
  "query": "VP of Engineering at Series B startups in San Francisco",
  "type": "premium",
  "insights": true,
  "profile_scoring": true,
  "high_freshness": true,
  "show_emails": true,
  "limit": 10
}

Credit Costs

Credits are charged based on features used per request. Pagination within cached results is free.

Rate Limits

• Per minute: 60 requests
• Per hour: 1,000 requests
• Monthly: Plan-dependent