--- type: "nimble-use-case" name: "account-enrichment" description: "Nimble's account enrichment agent populates a company record with the legal name, HQ, founding year, headcount, ownership, funding, current leaders and tech stack from the live web. Every value comes with the page it was read on and the date that page states." use_when: "Filling or refreshing firmographic fields for a list of companies, with a source on every value." not_for: "Emails or phone numbers, private financials, bulk refresh of millions of records." agent_id: "wsa_76fc85fc76ce46a4bcc2eec6194b5f8f" price_per_agent_run: { low: 0.025, medium: 0.1, high: 0.5 } updated: "2026-11-11" --- # AI web research agent for account enrichment (Nimble) Nimble's account enrichment agent populates a company record with the legal name, HQ, founding year, headcount, ownership, funding, current leaders and tech stack from the live web. Every value comes with the page it was read on and the date that page states. - Human page: https://www.nimbleway.com/use-cases/gtm/account-enrichment - Interactive demo: https://www.nimbleway.com/demos/account-enrichment - Agent context: https://www.nimbleway.com/use-cases/gtm/account-enrichment/account-enrichment-agent.md - Skill for agents: https://docs.nimbleway.com/skills/account-enrichment/SKILL.md - Docs: https://docs.nimbleway.com/nimble-sdk/web-search-agents/overview ## 1. Choose this when - You have a list of domains and need the same fields for each, read from current pages rather than a database snapshot. - You need to see where each value came from: every field carries a source URL and the date the source states. - You want tech stack from evidence, such as job posts and integrations pages. - You want JSON your CRM, spreadsheet or pipeline can take as it is, with unverified fields listed instead of guessed. ## 2. Use something else when - You need emails or phone numbers. The agent does not construct contact details. - You need private revenue or audited financials. Public pages do not have them. - You need people at the company by role. Use the decision-maker agent. ## 3. Minimal request One run of the ready-made account enrichment agent (`wsa_76fc85fc76ce46a4bcc2eec6194b5f8f`). ```bash AGENT_ID=wsa_76fc85fc76ce46a4bcc2eec6194b5f8f # Nimble's account enrichment agent curl -X POST "https://sdk.nimbleway.com/v2/agents/$AGENT_ID/runs" \ --header "Authorization: Bearer $NIMBLE_API_KEY" \ --header 'Content-Type: application/json' \ --data '{"input":"Enrich the company at gong.io","effort":"high"}' # poll until is_active is false, then read the result curl "https://sdk.nimbleway.com/v2/agents/$AGENT_ID/runs/$RUN_ID" --header "Authorization: Bearer $NIMBLE_API_KEY" curl "https://sdk.nimbleway.com/v2/agents/$AGENT_ID/runs/$RUN_ID/result" --header "Authorization: Bearer $NIMBLE_API_KEY" ``` The record is in `output.content`. Firmographic fields are `{ value, as_of, source_url }`; `leadership` is a list of `{ name, role, title, as_of, source_url, linkedin_url }`; `tech_stack` is a list of `{ tool, category, evidence_url }`; `last_round` holds the round, amount, currency, date and lead investors; `missing_fields` lists what no page could verify. ## 4. Use it from an agent Paste this into Claude, ChatGPT, Codex, Cursor or Grok: ```text Be my CRM enrichment assistant using Nimble. Follow https://docs.nimbleway.com/skills/account-enrichment/SKILL.md. ``` The full skill is in section 7 below. - Frameworks: `@nimble-way/mastra`, `langchain-nimble`, `@nimble-way/ai-sdk`, `llama-index-tools-nimble`, `nimble_python`, `@nimble-way/nimble-js`. ## 5. Example runs ### Ramp (recorded 2026-11-11) Input: Enrich the company at ramp.com Run record: high effort, 3.3 minutes, 11 pages cited. - Legal name: Ramp Business Corporation (2025-06-09) https://opengovny.com/corporation/5566048 - Domain: ramp.com - Industry: Fintech — financial operations platform (corporate cards, expense management, bill-pay software) (2026-10-22) https://en.wikipedia.org/wiki/Ramp_(company) - HQ city: New York (2026-11-11) https://ramp.com/about-us - HQ country: United States (2026-10-15) https://www.linkedin.com/company/ramp/ - Founded: 2019 (2025-06-09) https://opengovny.com/corporation/5566048 - Employees: 1,001–5,000 (LinkedIn company page shows 2,900 on the profile) (2026-11-11) https://www.linkedin.com/company/ramp/ - Ownership: Private (privately held, no ticker); venture-backed — investors include Founders Fund, Thrive Capital, Khosla Ventures, Sequoia, Greylock, Stripe, Goldman Sachs, Coatue and Redpoint (2026-11-11) https://www.linkedin.com/company/ramp/ - Total raised: Approximately $5 billion in total equity financing (2026-10-15) https://www.prnewswire.com/news-releases/ramp-reaches-60-billion-valuation-as-annualized-revenue-tops-2-billion-302909315.html - Last round: Series G, $1.9B, 2026-10-15, Dragoneer Investment Group, Thrive Capital, Founders Fund, Sands Capital, Avenir Growth https://www.prnewswire.com/news-releases/ramp-reaches-60-billion-valuation-as-annualized-revenue-tops-2-billion-302909315.html - Leadership: Eric Glyman (Co-Founder & Co-CEO, Ramp); Karim Atiyeh (Co-Founder & Co-CEO, Ramp); Will Petrie (CFO) - Tech stack: TypeScript, React, Vite, Ryu (Ramp's in-house design system), Python, AWS, Snowflake ### Attio (recorded 2026-11-11) Input: Enrich the company at attio.com Run record: high effort, 2 minutes, 8 pages cited. - Legal name: Attio Limited (incorporated in England and Wales; US entity: Attio Inc., Delaware) (November 2026) https://attio.com/legal/services-agreement - Domain: attio.com - Industry: Software Development (CRM software) (2026-11-11) https://www.linkedin.com/company/attio/ - HQ city: London (November 2026) https://attio.com/legal/services-agreement - HQ country: United Kingdom (November 2026) https://attio.com/legal/services-agreement - Founded: 2019 (2026-11-11) https://www.linkedin.com/company/attio/ - Employees: 51-200 employees (LinkedIn range; 205 listed employees on profile) (2026-11-11) https://www.linkedin.com/company/attio/ - Ownership: Privately held, VC-backed (GV, Redpoint Ventures, Balderton Capital, Point Nine, 01A) (2026-11-11) https://www.linkedin.com/company/attio/ - Total raised: $116 million (2025-08-26) https://prnewswire.com/news-releases/attio-raises-52m-series-b-to-scale-the-first-ai-native-crm-for-go-to-market-builders-302538357.html - Last round: Series B, $52M, 2025-08-26, GV (Google Ventures) https://prnewswire.com/news-releases/attio-raises-52m-series-b-to-scale-the-first-ai-native-crm-for-go-to-market-builders-302538357.html - Leadership: Nicolas Sharp (Co-Founder and CEO (current position per LinkedIn)); Alexander Christie (Co-Founder & CTO (current position per LinkedIn)) - Tech stack: Terraform, Kubernetes, GCP (Google Cloud Platform), Docker, Datadog, Splunk, TypeScript, Go, Python, Rust, OpenAI ### Gong (recorded 2026-11-11) Input: Enrich the company at gong.io Run record: high effort, 2.4 minutes, 8 pages cited. - Legal name: Gong.io Ltd. (2026) https://gong.io/about - Domain: gong.io - Industry: Revenue intelligence / revenue AI software (Revenue AI Operating System for sales, RevOps and customer success teams) (2026) https://gong.io/about - HQ city: San Francisco, California (201 Spear St, 13th Floor, San Francisco, CA 94105) https://www.linkedin.com/company/gong-io - HQ country: United States https://www.linkedin.com/company/gong-io - Founded: 2015 https://www.linkedin.com/company/gong-io - Employees: 1,001-5,000 employees (LinkedIn company size range; 2,112 members shown on LinkedIn) https://www.linkedin.com/company/gong-io - Ownership: Private, VC-backed (company states it is a private company; LinkedIn lists it as Privately Held; last priced round was Series E in June 2021) (2021-06-03) https://gong.io/press/gong-raises-250-million-in-series-e-funding-at-7-25-billion-valuation - Total raised: $584 million (2021-06-03) https://gong.io/press/gong-raises-250-million-in-series-e-funding-at-7-25-billion-valuation - Last round: Series E, $250M, 2021-06-03, Franklin Templeton https://gong.io/press/gong-raises-250-million-in-series-e-funding-at-7-25-billion-valuation - Leadership: Amit Bendov (CEO & Co-Founder); Eilon Reshef (Chief Product Officer & Co-Founder at Gong); Tim Ritters (Chief Financial Officer); Shane Evans (Chief Revenue Architect, Gong); Sandi Kochhar (Chief People Officer); Joe FitzGerald (Chief Legal Officer); Liron Shtraichman (Chief Research & Development (R&D) Officer); Guy Havusha (Chief Information & Security Officer (CISO)) - Tech stack: Greenhouse, Python, Salesforce, HubSpot, Calendly, Crossbeam, DealHub, Trumpet, Zendesk, Kubernetes, Cloudflare, GitHub, Veracode ### Notion (recorded 2026-11-11) Input: Enrich the company at notion.so Run record: high effort, 4 minutes, 13 pages cited. - Legal name: Notion Labs, Inc. (2026) https://notion.com/about - Domain: notion.so - Industry: Software Development (productivity / all-in-one workspace software) https://linkedin.com/company/notionhq - HQ city: San Francisco, California (2025-07-15) https://bizprofile.net/ca/san-francisco/notion-labs-inc - HQ country: United States https://siliconvalleyinvestclub.com/id/companies/notion - Founded: 2013 https://en.wikipedia.org/wiki/Notion_(productivity_software) - Employees: 501-1,000 employees (LinkedIn company-page range) https://linkedin.com/company/notionhq - Ownership: Privately held, venture/institutional-backed — investors include Sequoia Capital, Index Ventures, GIC and Coatue; valued at $11B in the January 2026 tender; no public ticker (2026-01-26) https://notion.com/blog/gic-sequoia-index-purchase-notion-shares - Total raised: $343.2 million raised across 3 rounds (as stated by Latka; other trackers state $330M-$340M) (2026-10-07) https://getlatka.com/companies/notion - Last round: Private tender offer (secondary purchase of employee shares, not primary capital), $270M, 2026-01-26, GIC, Sequoia Capital, Index Ventures https://notion.com/blog/gic-sequoia-index-purchase-notion-shares - Leadership: Ivan Zhao (Co-Founder & CEO); Simon Last (Co-Founder); Akshay Kothari (Co-Founder (current position on LinkedIn)); Max Schoening (CTO); Rama Katkar (CFO); Erica Anderson (Chief Revenue Officer); Lena Waters (Chief Marketing Officer); Madhu Muthukumar (Chief Product Officer) - Tech stack: TypeScript, JavaScript, React, Python, Java, Node.js, SQL, AWS, Google Cloud Platform (GCP), Microsoft Azure, Kubernetes, Git, Salesforce, LLMs / RAG / prompt engineering - Not verified: employee_count (official company-stated headcount — Notion has never published an exact figure; only the LinkedIn 501-1,000 range and third-party estimates ~1,000 in Sep 2025 / ~3,160 in Jun 2026 are available) ## 6. Questions ### What is an AI account enrichment agent? A research agent that takes a company, usually by its domain, reads current public pages about it and returns a structured record: legal name, HQ, founding year, employee range, ownership, funding, current leaders and tech stack. Nimble's agent returns each value with the URL it was read on and the date that page states, and lists the fields no page could verify. ### How is this different from an enrichment database? A database returns what it stored, refreshed on its own schedule. The agent reads the live pages when you ask, so each value carries the page and the date behind it and you can judge how current it is. When no page states a value, the field comes back empty and is listed in missing_fields rather than estimated. ### Where does the agent find a company’s tech stack? From evidence that the company uses a tool: job posts that list required tools, integrations pages, and customer case studies published by the vendor. Tools the company sells are not counted as its stack. Each tool comes back with the URL of the evidence. ### Can I run account enrichment from Claude, ChatGPT or my own agent? Yes. Paste this prompt into Claude, ChatGPT, Codex, Cursor or Grok: "Be my CRM enrichment assistant using Nimble. Follow https://docs.nimbleway.com/skills/account-enrichment/SKILL.md." The skill asks what you need, connects Nimble if it isn't connected yet, and runs the agent. Developers can call the same ready-made agent by its ID from the API, the Python or TypeScript SDK, or from Mastra, LangChain, the Vercel AI SDK and LlamaIndex. ## 7. Skill The skill the prompt points to, as published at https://docs.nimbleway.com/skills/account-enrichment/SKILL.md (copied 2026-11-11). ### Account enrichment Fills a CRM record for a company from its domain: firmographics, current leadership, tech stack and funding, every field cited or left empty. #### What the user gets One record per company: legal name, HQ, founded, employees, industry, ownership, leadership, tech stack and funding. Each field has a source, or is left empty with a reason. #### Agent template This skill runs the Nimble gallery template `account-enrichment`. The agent's instructions, sources and output schema live in the template. Do not rewrite them in the run. Effort: this skill runs the template at medium effort (about 1-3 minutes), even when the template's default effort is different. That is on purpose. If the user asks for a specific effort (low, medium, high or x-high), run that one instead. #### Flow 1. Interview the user. 2. Connect Nimble if it is not connected already. 3. Run the agent and show the result. 4. Offer next steps. Ask every question with the host's question UI, never as plain text: `AskUserQuestion` in Claude Code, ChatGPT's interactive question form (clickable choices and a Submit button) in ChatGPT. ##### Get the inputs | Input | Required | Ask for | If skipped | | - | - | - | - | | Company | yes | Its domain | - | | Extra fields | no | Anything beyond the default fields, for example "SOC 2 status" or "uses Salesforce" | standard fields only | * Skip anything the user already said in this conversation or in their request, and ask only for what is missing. Do not fill in answers from earlier conversations, with one exception: a stable fact about the user, such as what they sell, may be offered as the first option, labeled "(from before)". Never carry over the subject of the request, dates or rules from an earlier conversation. * Ask the required inputs first, in table order, one question at a time. Then ask the optional ones together, in one question-tool call, as many as the tool takes per call (a second call for the rest). * Put the whole question in the question tool, once. Never restate or introduce a question in text. Between questions, acknowledge the answer in a few words, and before the first question, at most one short sentence that acknowledges your role (the user's, if they named one), such as "Sure, I'll use Nimble to be your CRM enrichment assistant." Talk in plain, friendly words: never say "input", "required" or "optional", and use no field names. Use plain text only if the host has no question tool, in one short line. * Write 2-3 likely answers as options, based on what the user has already said. Fixed examples in the table are a guide to what the answer contains, not a list to copy. Offer only answers Nimble can act on with public web data, never ones that need the user's own data, such as their customers or CRM. * If you cannot guess likely answers, such as a company name or an industry, still use the tool and offer two or three examples as options, written as answers the user could really give. If the user picks one, use it as their answer. * Every question must let the user type their own answer, optional ones too. Claude's question UI adds its own free-text choice ("Something else" in Claude, "Other" in Claude Code), so add none there. ChatGPT's form does not, so add "Other" as the last option of every question, even one whose options look complete, opening a text field with a placeholder. In any other host, add "Other" unless you know its UI already has a free-text choice. Keep Skip and Other, when you add them, even if you must drop an answer to fit the option limit. * Use single-select only. For a range, such as a size or a date, offer a few options that do not overlap. * Optional questions can be skipped, and a skipped answer uses the "If skipped" value. Claude's question UI has its own Skip button, so add no Skip option there. In other hosts, include a Skip option that states the default in its label, for example "Skip (any location)", not in the question. When you acknowledge a skipped question, say the default you will use, for example "Any location.", not that it was skipped. Do not offer a "skip all" option. If the user asks to skip the remaining questions, use the "If skipped" values for all of them. * Do not ask about effort. Build the agent input from this pattern: ``` Enrich the company at ``` Example: `Enrich the company at attio.com` * One company at a time. If the user names several, run the first and offer the rest next. * Extra fields go in the input as one line each, for example: "Also return: SOC 2 status (yes/no, with source)." ##### Connect Nimble Do this after the interview. Until the questions are done, say nothing about connecting Nimble. Check the connection. If a Nimble tool is listed in this session, call `nimble_agents_list`. If the CLI is available, run `nimble --client-source nimble-agent-skills agents list`. A normal response from either, even an empty list, means the user is connected: say nothing and go on. If neither check succeeds (no tool is listed and the CLI is missing or fails, or a check returns an auth error or an authorization link), connect it in these steps. Your message is only these steps: no summary of what the user told you, no mention of the skill, the template or the workflow, and no plan for what happens after connecting (that belongs in the run notice). 1. Tell the user, in two short sentences at most: "Nimble searches the live web and links where every answer came from. Connecting takes one approval, and if you're new to Nimble you start with a free trial." 2. Connect it in the first of these ways that works, and stop at the first one that does: * Show the host's card to connect or install Nimble from nimbleway.com: the Connect card for the connector in Claude, the install card for the plugin in ChatGPT. * If you can't show a card and a check returned an authorization link, show that link to the user. * If you can do neither but can run commands, follow [https://docs.nimbleway.com/agent-onboarding.md](https://docs.nimbleway.com/agent-onboarding.md) and involve the user only for the sign-up and the approval. With only the CLI and no `NIMBLE_API_KEY`, an approval is not enough: follow the page's API key fallback and have the user set the key in their environment, never in the chat. That page is written for you: never link it or its steps to the user. * If none of these works, tell the user how to find Nimble in this app, with the clicks: in ChatGPT, open Plugins in the sidebar, search for Nimble, pick the one described as "Expert web search for agents" and install it; in Claude, open Customize, then Connectors, then Nimble, then Connect. Never send the user to nimbleway.com for this. 3. Ask with the question tool, with the options "I've connected Nimble" and "Not now". If the host has no question tool, end the message by asking them to reply "connected", and add nothing after that. Never ask the user to paste an API key into the chat. When they confirm, check the connection again and go on without comment. If it is still not connected, say so in one line and ask again. If they pick "Not now", say the skill needs Nimble and stop. Show no research until Nimble is connected, not from your own search and not from memory. ##### Run the agent Do not comment on the connection check or on creating or reusing the agent. The first thing the user hears is step 2. 1. Create the agent from the template: `nimble_agents_create(template="account-enrichment", agent_name="account-enrichment")`. A 409 "already exists" means it was created before, so go on. If the template is not found, tell the user it is unavailable and stop. Do not substitute another template. 2. Tell the user once, in one line, what you are researching and how long it takes at the effort you will use: "Researching \. Expected to take \