메뉴
Search for companies or people using Fiber AI. Use when the user wants to find companies by criteria (location, industry, size, tech stack) or search for individual prospects.
Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.
SOC 직업 분류 기준
| name | search |
| description | Search for companies or people using Fiber AI. Use when the user wants to find companies by criteria (location, industry, size, tech stack) or search for individual prospects. |
| user-invocable | true |
| argument-hint | <search query in natural language> |
Search for companies or people matching criteria using Fiber AI MCP tools.
/fiber:enrich instead/fiber:audience instead/fiber:sdk-ts or /fiber:sdk-py insteadfiber-ai-v2)The V2 MCP has direct tools for search. Tool names follow the pattern {operationId}_tool:
companySearch_toolpeopleSearch_toolfiber-ai-core)get_endpoint_details_full("companySearch") or get_endpoint_details_full("peopleSearch") to get the exact parameter schemacall_operation("companySearch", {"body": {...}}) or call_operation("peopleSearch", {"body": {...}}) with the correct parametersFiber's search filter schemas are large and evolve across versions. Never hardcode or guess field names. Discover the current schema at runtime using one of these methods (in preference order):
@fiberai/sdk exports typed request/response models. Check node_modules/@fiberai/sdk types or use IDE autocomplete. Install: npm install @fiberai/sdk.fiberai ships typed model classes (e.g. fiberai.models.CompanySearchBody). Inspect with help() or IDE. Install: pip install fiberai.get_endpoint_details_full("companySearch") or get_endpoint_details_full("peopleSearch") on the Core MCP to get the exact current schema with all filter fields.https://api.fiber.ai/ai-docs/companySearch.md / https://api.fiber.ai/ai-docs/peopleSearch.mdhttps://api.fiber.ai/llms.txthttps://api.fiber.ai/docs/https://github.com/fiber-ai/open-fiberKey structural notes (stable across versions):
searchParams object (not filters)cursor (not page) and pageSizeapiKey is a required field in the request bodyvp, director, c-level, manager, etc.) — not regexlowerBound/upperBound semantics, not custom operatorschargeInfo with exact credits chargedExample body structure (verify exact field names via methods above):
{
"apiKey": "...",
"searchParams": { ... },
"pageSize": 25,
"cursor": null
}
If the user gives freeform intent (e.g., "VPs of Engineering at fintech startups in NYC") and you cannot extract clean filters, use Fiber's NL-to-query endpoints:
textToCompanySearch — translates free text into resolved companySearchParams with canonicalized identifierstextToProfileSearch — translates free text into resolved profileSearchParams with typed seniority, job status, and entity-resolved companiesThese endpoints save the entire filter-construction step. They resolve company names to LinkedIn org IDs and map role descriptions to typed seniority levels automatically.
Every API call requires apiKey in the request body. Check https://api.fiber.ai/docs/ for the exact authentication format for each endpoint. If the user hasn't configured authentication, direct them to run /fiber:setup.
/fiber:audienceSearch charges credits per result found (not per page). Check the current pricing via the response's chargeInfo object, at https://api.fiber.ai/ai-docs/companySearch.md / https://api.fiber.ai/ai-docs/peopleSearch.md, or via the llms://fiber.ai/llms.txt MCP resource — costs may change.
Accept: text/markdown to https://api.fiber.ai/openapi.json to get the agent-friendly index at the same URL (Stripe pattern)./fiber:setup/fiber:setup for API key configurationRun a reproducible, pre-registered benchmark comparing Fiber AI to a competing data provider (People Data Labs, Apollo, Clearbit, ZoomInfo, Coresignal, or any other API the user has credentials for) on the user's own sample. Reports honest side-by-side numbers including where Fiber underperformed. Use this when the user says "compare Fiber to <vendor>", "bake-off", "benchmark data quality", "which provider is better for my segment", "test data providers", "evaluate Fiber AI", or "should I switch from <vendor> to Fiber".
Enrich a list of LinkedIn URLs with work emails, personal emails, and phone numbers using Fiber AI. Use this when the user says "enrich these LinkedIn URLs", "I have a CSV of people", "bulk reveal emails for these profiles", "get contact info for this list", "decorate these contacts with phone/email", or otherwise hands over a set of LinkedIn identifiers and wants contact data back.
Enrich companies or contacts with emails, phone numbers, and detailed profiles using Fiber AI. Use when the user wants to reveal contact information or get more details about a company or person.
Find prospects by role + company criteria and enrich a shortlist with contact details using Fiber AI. Use this when the user says "find VPs of Engineering at fintech startups", "CMOs at Series B SaaS", "heads of sales at X companies", "find me 20 CTOs in healthcare", "prospect by role + company", "sales target research", or otherwise wants to discover people by role and immediately get emails / phones back.
Find lookalike companies from one or more seed accounts using Fiber AI. Use this when the user says "find companies like <seed>", "competitors of <seed>", "similar to <seed>", "lookalike accounts", "companies that look like our best customers", "ABM lookalikes", "seed-based company search", or otherwise asks to expand a known account into a larger candidate list.
Show what Fiber AI can do and how to use it. Use when the user asks about Fiber AI capabilities, available commands, or needs general guidance.