# Account Brief > Walk into any meeting knowing the account. Give a company and the meeting; Nimble agents read the live web and write a one-page brief with what's happening now, who you'll meet and what to ask, every line sourced and dated. Interactive demo: https://www.nimbleway.com/demos/account-brief ## Walk into any meeting knowing the account Give a company and the meeting. Nimble agents read the live web and write a one-page brief: what's happening now, who you'll meet and what to ask. ## Examples to try - "Prep me for a first call with Snowflake" - Ramp, Snowflake, Figma, HubSpot ## Set it up in your agent The Nimble skill for this use case: https://docs.nimbleway.com/skills/account-brief/SKILL.md Prompt to paste into your agent: ```text Be my meeting preparation assistant using Nimble. Follow https://docs.nimbleway.com/skills/account-brief/SKILL.md. ``` Get an API key: https://online.nimbleway.com/signup ## The most accurate & efficient web search for GTM data We asked Opus 5.5 hundreds of GTM questions about companies, people and industry news. It ran once with each web search provider, and every answer was checked against a hand-verified ground truth. We tracked accuracy and the total cost of getting there, including search fees and tokens. Benchmark code and data: https://github.com/Nimbleway/gtm-enrichment-benchmark ## What you get A one-page brief on one company before a sales meeting, researched on the live web: what the company does, current priorities and challenges with evidence, what changed in the last 90 days, key people and talking points. Every item is dated and links to the page it came from, and fields the agent couldn't date or source are listed as missing. ## Sections You pick the kind of meeting (first call, demo / evaluation, renewal / expansion, or partnership). The demo then researches the brief in six parallel runs and lays it out in eight sections. | Section | What it holds | |---|---| | At a glance | A short summary, what they do, who they sell to, HQ, founded, employees, ownership and funding, revenue | | What's happening now | Dated events (launches, earnings, partnerships and more), each with why it matters for the meeting | | Their priorities | What the company is focused on and the pressures it faces, with evidence | | People in the room | Key people with title, time in role, background and a recent item, each sourced | | Today's tools and competitors | Tools they use and their competitors, plus a line on what you sell when you give it | | Where you could fit | Angles for what you sell, each with a reason and a source. Shown only when you say what you sell | | Questions to ask | Questions for the meeting, each with the reason and a source | | Watch-outs | Dated topics to handle with care, each with the detail, a source and how to handle it in the room | ## Inputs | Input | Required | Example | |---|---|---| | Company | Yes | Name and domain, for example Attio (attio.com) | | What you sell | No | One line. It sharpens the talking points | | People in the meeting | No | Names and titles, if you want a line on each | ## Run it yourself The Nimble skill for this use case runs the Nimble agent template `account-brief`: https://docs.nimbleway.com/skills/account-brief/SKILL.md - Create the agent once: `nimble_agents_create(template="account-brief", agent_name="account-brief")` - Run it: `nimble_agents_run(agent_name="account-brief", input="Prepare me for a meeting with (). I sell . The meeting is .", effort="medium")` - Effort: the skill runs the template at medium effort (about 1-3 minutes), even when the template's own default is different. Ask for low, medium, high or x-high to run another. Always include the domain, so the agent doesn't drift to a company with a similar name. If you name the people in the meeting, add them to the input. Challenges come from reviews, press and job posts, so a short list can be real. Recent changes older than 90 days should not be there. ## When to use it, and when not to Use it for "prep me for a meeting with", "account research on", "brief me on [company] before my call", "account brief" or "what should I know before meeting" requests. - For an investment memo, use Due Diligence: https://www.nimbleway.com/demos/due-diligence.md - For a list of changes at an account since a date, use Account Signals: https://www.nimbleway.com/demos/account-signals.md - For a CRM record of firmographics and tech stack, use Account Enrichment: https://www.nimbleway.com/demos/account-enrichment.md - To map who to talk to beyond the meeting, use Decision-Maker Identification: https://www.nimbleway.com/demos/decision-maker-identification.md ## Sample result From the demo's recorded run on Snowflake (snowflake.com), first-call meeting, At a glance section, high effort, recorded 2026-11-10. The agent considered 78 pages and read 24. | Field | Value | |---|---| | HQ | Menlo Park, California, U.S. (relocated back to California in 2025; founded in San Mateo) | | Founded | 2012 (founded July 23, 2012, in San Mateo, California, by Benoît Dageville, Thierry Cruanes and Marcin Żukowski) | | Employees | 9,394 total employees (company, Q2 FY2027 headcount by function), as of 2026-07-31 | | Ownership | Public — NYSE: SNOW (Class A); IPO September 2020 | | Customers | 14,554 total customers as of July 31, 2026, including 829 Forbes Global 2000 companies and 828 customers with over $1M trailing-12-month product revenue | | Revenue | FY2026 total revenue $4.68 billion, of which product revenue $4,472 million, up 29% Y/Y (company, Q2 FY2027 investor presentation) | Sources include https://en.wikipedia.org/wiki/Snowflake_Inc., https://investors.snowflake.com/files/doc_financials/2027/q2/Q2-FY2027-Investor-Presentation_vF.pdf and https://www.snowflake.com/en/product/platform/ The run listed four missing fields: no page gave a date for the HQ, founding, listing or product description facts. The other sections return dated, sourced items the same way. --- # Account Discovery > Describe your ideal customer in plain English. Nimble agents search the live web for companies that fit, check every criterion against a source, and return a list you can export to your CRM. Interactive demo: https://www.nimbleway.com/demos/account-discovery ## Find every company that fits your ICP Describe your ideal customer. Nimble agents search the live web for companies that match and check every criterion against a source. ## Examples to try - "Find Series A fintechs in the US hiring SDRs" - Series A–B US fintechs hiring their first SDRs, Vertical SaaS for dental clinics in North America, European logistics companies with 200–1,000 employees, AI startups that launched a public API in the last 6 months - Ramp, Vanta, Attio, Deel ## Set it up in your agent The Nimble skill for this use case: https://docs.nimbleway.com/skills/account-discovery/SKILL.md Prompt to paste into your agent: ```text Be my prospecting assistant using Nimble. Follow https://docs.nimbleway.com/skills/account-discovery/SKILL.md. ``` Get an API key: https://online.nimbleway.com/signup ## The most accurate & efficient web search for GTM data We asked Opus 5.5 hundreds of GTM questions about companies, people and industry news. It ran once with each web search provider, and every answer was checked against a hand-verified ground truth. We tracked accuracy and the total cost of getting there, including search fees and tokens. Benchmark code and data: https://github.com/Nimbleway/gtm-enrichment-benchmark ## What you get A list of companies that match an ideal customer profile (ICP), researched on the live web. Each row has the company name, website, why it fits, employees, HQ and a recent dated signal with its source. Every row is checked against each criterion: met, not confirmed or not met, with the evidence, a source URL and a date where the page gives one. ## Modes | Mode | You give | Criteria it checks | |---|---|---| | Describe your ICP | One sentence, for example "Vertical SaaS for dental clinics in North America" | Category, plus funding stage, company size, region and any extra clause (such as "hiring their first SDRs") pulled out of the sentence | | Find lookalikes | One company, for example Ramp | Same category as that company, sells to the same buyers | In both modes you can add must-haves of your own and set filters for region, company size and funding stage. The demo asks for 10 or 20 results and searches three angles in parallel: directories and lists, news and funding, and hiring and product signals. ## Inputs | Input | Required | Example | |---|---|---| | Ideal customer | Yes | Industry or product, geography and size, in one sentence | | Hard rules | No | For example "raised Series A or later since 2024" or "not already a customer" | ## Run it yourself The Nimble skill for this use case runs the Nimble agent template `account-discovery`: https://docs.nimbleway.com/skills/account-discovery/SKILL.md - Create the agent once: `nimble_agents_create(template="account-discovery", agent_name="account-discovery")` - Run it: `nimble_agents_run(agent_name="account-discovery", input="Find in with , that . Companies like .", effort="medium")` - Effort: the skill runs the template at medium effort (about 1-3 minutes), even when the template's own default is different. Ask for low, medium, high or x-high to run another. Turn each criterion into a clear test: "mid-market" is not a test, "200-2,000 employees" is. For a broad profile (a whole country or a big vertical), split it into slices by region or sub-vertical, one run per slice in parallel, then merge and dedupe by website. Don't ask for a number of rows; the agent returns every company it can verify. Headcount for private companies is often a range, so accept a range. Fewer rows than expected is a signal to run another slice, not a finished list. ## When to use it, and when not to Use it for "find companies like", "build a prospect list", "ICP list", "find accounts that match", "target account list" or "companies that fit our ICP" requests. - To research one known company, use Account Brief: https://www.nimbleway.com/demos/account-brief.md - To find investment targets, use Deal Sourcing: https://www.nimbleway.com/demos/deal-sourcing.md - To fill firmographics and tech stack for the companies you found, use Account Enrichment: https://www.nimbleway.com/demos/account-enrichment.md - To find the people to contact at them, use Decision-Maker Identification: https://www.nimbleway.com/demos/decision-maker-identification.md ## Sample result From the demo's recorded run for "Vertical SaaS for dental clinics in North America", Describe your ICP mode, 10 results requested. The run checked 9 companies against two criteria (vertical SaaS for dental clinics, HQ in North America), and all 9 met both. Five of the rows: | Company | Website | HQ | Employees | |---|---|---|---| | Archy | archy.com | San Jose, California, USA | 138 (LinkedIn, 51-200 range); 130 per company blog | | Curve Dental | curvedental.com | Alpharetta, Georgia, USA | 214 (LinkedIn, 201-500 range) | | tab32 | tab32.com | Sacramento, California, USA | 50+ (company About page); 55 (LinkedIn) | | Planet DDS | planetdds.com | Irvine, California, USA | 440 | | ClearDent | cleardent.com | Burnaby, British Columbia, Canada | Not given | Sources include https://www.curvedental.com/, https://www.linkedin.com/company/archydental/ and https://www.archy.com/blog/weve-raised-50-million-to-bring-dental-practices-into-the-ai-era Each row also carries a one-line reason it fits and a dated signal, such as a funding round or a product launch, with its source. --- # Account & Lead Enrichment > Enrich any account or lead from the live web: firmographics, tech stack, funding, hiring and any custom field you ask for, every value with a source and date, as clean JSON or CSV for your CRM or product. Interactive demo: https://www.nimbleway.com/demos/account-enrichment ## Enrich any account or lead from the live web Give a company or a person. Nimble agents search the live web for firmographics, tech stack, funding, hiring and any custom field you ask for. ## Examples to try - "Enrich Ramp with firmographics and tech stack" - Ramp, Attio, Gong, Notion, Eric Glyman · Ramp, Ivan Zhao · Notion, Amit Bendov · Gong ## Set it up in your agent The Nimble skill for this use case: https://docs.nimbleway.com/skills/account-enrichment/SKILL.md Prompt to paste into your agent: ```text Be my CRM enrichment assistant using Nimble. Follow https://docs.nimbleway.com/skills/account-enrichment/SKILL.md. ``` Get an API key: https://online.nimbleway.com/signup ## The most accurate & efficient web search for GTM data We asked Opus 5.5 hundreds of GTM questions about companies, people and industry news. It ran once with each web search provider, and every answer was checked against a hand-verified ground truth. We tracked accuracy and the total cost of getting there, including search fees and tokens. Benchmark code and data: https://github.com/Nimbleway/gtm-enrichment-benchmark ## What you get One enriched record per company, researched on the live web: legal name, HQ, founded, employees, industry, ownership, leadership, tech stack and funding. Each field carries its value, a source URL and an as-of date, or is left empty with a reason (`missing_fields`). Each section of the record also gets a confidence grade. ## Modes The demo enriches either an account (a company) or a lead (a person at a company). You pick the fields, and each field group is researched in its own run. | Mode | Default fields | Optional fields | |---|---|---| | An account | Firmographics (industry, size, HQ, founded, ownership, funding, revenue), Tech stack (tools they use, with the evidence), Funding (every round, amount and lead investors), Headcount & hiring (employee count, its trend and open roles by function) | Offices & regions, Pricing & sales motion, Customers & segments | | A lead | Profile (title, company, location, tenure, public profiles), Work history (previous companies and titles), Company (industry, size, funding and the latest company news) | Education | Both modes take up to three custom fields in plain English, for example "Which CRM do they use?" or "Are they SOC 2 compliant?". Each comes back with a value, a source and a date. ## Inputs | Input | Required | Example | |---|---|---| | Companies | Yes | Domains, pasted or as a CSV / sheet column, for example attio.com | | Extra fields | No | Anything beyond the default fields, for example "SOC 2 status" or "uses Salesforce" | | Record ID | No | Your CRM ID per company, so rows can be matched back | ## Run it yourself The Nimble skill for this use case runs the Nimble agent template `account-enrichment`: https://docs.nimbleway.com/skills/account-enrichment/SKILL.md - Create the agent once: `nimble_agents_create(template="account-enrichment", agent_name="account-enrichment")` - Run it: `nimble_agents_run(agent_name="account-enrichment", input="Enrich the company at ", effort="medium")` - Effort: the skill runs the template at medium effort (about 1-3 minutes), even when the template's own default is different. Ask for low, medium, high or x-high to run another. One domain per run. For a list, start one run per domain in parallel; many domains in one run come back thin. Extra fields go in the input as one line each, for example "Also return: SOC 2 status (yes/no, with source)." Tech stack means tools the company uses, not tools it sells. A missing CFO or CRO is often real, so check `missing_fields` before calling it a gap. ## When to use it, and when not to Use it for "enrich these domains", "fill my CRM", "firmographics for", "enrich this account list", "what tech stack does" or "company data for" requests. - To find who holds a specific role, use Decision-Maker Identification: https://www.nimbleway.com/demos/decision-maker-identification.md - For a brief before a meeting, use Account Brief: https://www.nimbleway.com/demos/account-brief.md - To build the list of companies to enrich, use Account Discovery: https://www.nimbleway.com/demos/account-discovery.md ## Sample result From the demo's recorded run on Ramp (ramp.com), account mode, Firmographics section, low effort, recorded 2026-11-09. The agent ran 3 searches and 14 extracts and graded the section medium confidence, with no missing fields. | Field | Value | As of | |---|---|---| | Legal name | Ramp Business Corporation | 2026-11-09 | | HQ | New York, New York, United States | 2026-11-09 | | Founded | 2019 | 2026-11-09 | | Employees | 1,001-5,000 | 2026-11-09 | | Total funding | $5.07B (Tracxn) | 2026-10-22 | | Last round | Series G, $1.85B, October 2026, at ~$60B valuation | 2026-10-22 | Sources include https://www.linkedin.com/company/ramp, https://en.wikipedia.org/wiki/Ramp_(company) and https://crunchbase.com/organization/ramp-financial The other sections return their own fields the same way: tools grouped by category with the evidence for each, funding rounds with dates and lead investors, and headcount with open roles by function. --- # Account Signals > Give a watchlist of accounts. Nimble agents check the live web for funding, new leaders, hiring surges and launches, then rank who to call first and what to say. Every signal dated and sourced. Interactive demo: https://www.nimbleway.com/demos/account-signals ## Know which accounts to call this week Give a watchlist of accounts. Nimble agents check the live web for funding, new leaders, hiring and launches, and rank who to call first. ## Examples to try - "Which of my accounts raised money or hired a new VP?" ## Set it up in your agent The Nimble skill for this use case: https://docs.nimbleway.com/skills/account-signals/SKILL.md Prompt to paste into your agent: ```text Be my account monitoring assistant using Nimble. Follow https://docs.nimbleway.com/skills/account-signals/SKILL.md. ``` Get an API key: https://online.nimbleway.com/signup ## The most accurate & efficient web search for GTM data We asked Opus 5.5 hundreds of GTM questions about companies, people and industry news. It ran once with each web search provider, and every answer was checked against a hand-verified ground truth. We tracked accuracy and the total cost of getting there, including search fees and tokens. Benchmark code and data: https://github.com/Nimbleway/gtm-enrichment-benchmark ## What you get A list of what changed at an account since a date, researched on the live web: funding, executive moves, launches, pricing changes and hiring pushes. Only events dated after the since date are kept, each with its date, type, summary and source. If nothing changed, the result is an empty list plus the sources the agent checked. The demo runs one agent per account on a watchlist of up to five, then ranks the accounts by heat (0-100) with the reason for the score. Each signal also carries a strength from 1 to 3 and a suggested play, and each account gets a top play: the role to contact, what to say and why now. Events dated before the window, undated events and re-reported old news are dropped. ## Signal types | Signal | What it covers | On by default in the demo | |---|---|---| | Funding | Rounds and debt raises | Yes | | Leadership hires | Executives and VPs who joined, moved or left | Yes | | Hiring surge | Many new roles in one function, or a new team | Yes | | Expansion | New offices, regions or markets | Yes | | Product launch | Launches and major releases | Yes | | M&A | Acquisitions and mergers, either side | Yes | | Partnerships | Partnerships and integrations announced | No | | Layoffs & restructuring | Job cuts and reorganisations | No | | Tech change | Tools adopted, replaced or migrated | No | | Strategy statements | What leadership says they are investing in | No | The demo's window is 30, 90 or 180 days (default 90), and an optional "What do you sell?" line scores the ranking for your product. ## Inputs | Input | Required | Example | If skipped | |---|---|---|---| | Accounts | Yes | Names and domains, or a list | - | | Since | No | A date in YYYY-MM-DD | 30 days ago | | Signal types | No | "only exec hires and funding" | All | ## Run it yourself The Nimble skill for this use case runs the Nimble agent template `account-signals`: https://docs.nimbleway.com/skills/account-signals/SKILL.md - Create the agent once: `nimble_agents_create(template="account-signals", agent_name="account-signals")` - Run it: `nimble_agents_run(agent_name="account-signals", input="Check () for changes since ", effort="medium")` - Effort: the skill runs the template at medium effort (about 1-3 minutes), even when the template's own default is different. Ask for low, medium, high or x-high to run another. One account per run. For a list, run one per account in parallel. Always pass an explicit date, and keep only events strictly after it: events on the since date itself can slip in. To limit signal types, add one line to the input. ## When to use it, and when not to Use it for "what changed at", "buying signals", "track my accounts", "any news on my accounts since", "trigger events" or "signals for" requests. - For a full company brief before a meeting, use Account Brief: https://www.nimbleway.com/demos/account-brief.md - For changes at investment holdings, use Portfolio Monitoring: https://www.nimbleway.com/demos/portfolio-monitoring.md - For the people to contact at an account, use Decision Maker Identification: https://www.nimbleway.com/demos/decision-maker-identification.md ## Sample result From the demo's recorded run on Mercury (mercury.com), all six default signal types, 90-day window, low effort, recorded 2026-11-09. The agent ran 7 searches and 34 extracts, graded the result high confidence and scored Mercury at heat 58. | Date | Type | Strength | Summary | |---|---|---|---| | 2026-11-05 | launch | 3 | Mercury launched Agent Controls, which lets customers see every AI agent connected to their account, govern what each can do, and disconnect them in seconds. | | 2026-11-04 | launch | 2 | Mercury redesigned Treasury with a recommended portfolio built around monthly expenses and U.S. Treasury ladders for longer-term yield. | | 2026-10-20 | launch | 3 | Mercury launched Mercury Payroll across all 50 states: W-2 and 1099 contractor payroll, health benefits and 401(k), with white-glove compliance support; free through end of 2026, then $100/month plus $50 per person paid. | | 2026-10-15 | launch | 2 | Mercury enabled moving money, paying recipients and categorizing transactions from ChatGPT or Claude via MCP, with nothing moving until approved. | | 2026-09-29 | launch | 2 | Mercury added real-time payments through its API, settling in seconds 24/7/365 including nights, weekends and holidays. | Top play: contact the VP of Finance or Head of People Operations, opening with the Mercury Payroll launch. Sources include https://mercury.com/releases and https://mercury.com/ --- # Credit Risk Agent > Assess the credit risk of any company. Nimble agents read filings, rating actions, debt news, court and lien records on the live web and return a risk call with a source for every finding. Interactive demo: https://www.nimbleway.com/demos/credit-risk ## Assess the credit risk of any company Research a specific company with Nimble agents that search deep across the live web, reading filings, rating actions, debt and bankruptcy news, court and lien records. ## Examples to try - "What is the credit risk of Carvana?" - Carvana, Peloton, Wayfair, AMC ## Set it up in your agent The Nimble skill for this use case: https://docs.nimbleway.com/skills/credit-risk/SKILL.md Prompt to paste into your agent: ```text Be my credit research assistant using Nimble. Follow https://docs.nimbleway.com/skills/credit-risk/SKILL.md. ``` Get an API key: https://online.nimbleway.com/signup ## The most accurate & efficient web search for credit research We asked Opus 5.5 credit questions about companies: latest financials, rating actions, debt events and distress signals. It ran once with each web search provider, and every answer was checked against a hand-verified ground truth. We tracked accuracy and the total cost of getting there, including search fees and tokens. Benchmark code and data: https://github.com/Nimbleway/credit-risk-benchmark ## What you get A dated, sourced credit picture of one company, researched on the live web: financials with their periods, ratings, debt events, payment behaviour, distress signals and ownership. It ends with a risk call (low, moderate, elevated or high) and the reasons behind it. The risk call is always shown with its reasons and sources, never alone. Fields the agent couldn't source are shown as missing. ## Modes The demo asks what your exposure to the company is. Each mode opens with a snapshot and a risk call for its own question, then researches its sections. | Mode | Risk question | Sections | Snapshot | |---|---|---|---| | Trade credit | Likelihood invoices go unpaid or are paid very late in the next 12 months | Payment record, Cash & liquidity, Distress signals, Who stands behind them, Recent news | Bill-to entity, ownership, payment record, liquidity, revenue, latest rating, plus suggested terms and what to check before you extend terms | | Lending & private credit | Likelihood the company cannot service or refinance its debt in the next 12 months | Leverage & coverage, Debt stack & maturities, Covenants & amendments, Sponsor & ownership, Distress signals | Total debt, leverage, interest coverage, liquidity, next maturity, sponsor / owner, plus how to approach a loan and questions for management | | Credit investing | Likelihood of default or a distressed exchange in the next 12 months | Rating actions, Catalyst calendar, Capital structure & recovery, Results & guidance, News flow | Ratings, total debt, next maturity, key catalyst, plus rating direction and what could improve or hurt the credit | | Supplier risk | Likelihood they cannot keep supplying normally in the next 12 months | Financial stability, Operations & disruption, Distress signals, Ownership changes, Sanctions & compliance | Ownership, financial health, operations, locations, plus a recommended action and what to watch for | | Quick credit check | Likelihood the company cannot meet its obligations in the next 12 months | Financial health, Credit ratings, Debt events, Distress signals, Payment behavior, Ownership & structure, News | Ownership, latest rating, revenue, profitability, total debt, liquidity, next maturity | Lending and credit investing also lay out dated maturities and catalysts as a calendar. ## Inputs | Input | Required | Example | If skipped | |---|---|---|---| | Company | Yes | The exact legal entity if it is a subsidiary, with a domain or ticker | - | | Purpose | No | Lending, trade credit or a supplier check. It sets what the reasons stress | General | ## Run it yourself The Nimble skill for this use case runs the Nimble agent template `credit-risk`: https://docs.nimbleway.com/skills/credit-risk/SKILL.md - Create the agent once: `nimble_agents_create(template="credit-risk", agent_name="credit-risk")` - Run it: `nimble_agents_run(agent_name="credit-risk", input="Assess the credit and counterparty risk of (). Purpose: . Registered in .", effort="medium")` - Effort: the skill runs the template at medium effort (about 1-3 minutes), even when the template's own default is different. Ask for low, medium, high or x-high to run another. One company per run. For a subsidiary, name the exact legal entity: the agent will not use the parent's numbers. Many mid-size companies have no public rating, and the run says so. For private companies, figures come from filed accounts or press, so check who stated them. ## When to use it, and when not to Use it to decide whether a company can pay: before extending trade credit, lending, buying its debt or relying on it as a supplier. - For a broader first-pass memo on a company, use Due Diligence: https://www.nimbleway.com/demos/due-diligence.md - For sanctions, PEP and adverse-media screening of a counterparty, use KYC Screening: https://www.nimbleway.com/demos/kyc-screening.md - For ongoing changes across companies you hold, use Portfolio Monitoring: https://www.nimbleway.com/demos/portfolio-monitoring.md ## Sample result From the demo's recorded run on Wayfair (wayfair.com), Quick credit check, company snapshot, high effort, recorded 2026-11-10. The agent ran 5 searches and 26 extracts and graded the snapshot high confidence. | Field | Value | |---|---| | Revenue | FY2025 total net revenue $12,457 million (up 5.1% YoY); nine months ended September 30, 2026 net revenue $9,870 million (up from $9,120 million a year earlier), with Q3 2026 up 9.7% YoY | | Latest rating | S&P Global Ratings upgraded Wayfair LLC to 'BB' from 'B+' (senior secured debt to 'BB+'), outlook stable, on August 11, 2026; Moody's upgraded Wayfair Inc.'s corporate family rating to B1 from B3, outlook stable, on September 22, 2026; Fitch upgraded Wayfair Inc./Wayfair LLC to 'BB-' from 'B', outlook stable, on May 13, 2026 | | Total debt | Long-term debt of $2,572 million as of September 30, 2026, down from $3,233 million at December 31, 2025 (senior secured notes due 2029/2030/2032/2034 plus $268 million of convertible notes remaining at June 30, 2026) | | Liquidity | Cash, cash equivalents and short-term investments of $1.1 billion and total liquidity of $1.5 billion, including availability under the revolving credit facility, as of September 30, 2026 (balance sheet: cash $1,038 million + short-term investments $60 million; $418 million available under the $500 million revolver at June 30, 2026) | | Next maturity | $229 million of convertible senior notes due September 2027 (the $39 million convertible maturity due August 2026 was repaid at maturity in Q3 2026); remaining senior secured notes mature 2029-2034 and the $500 million revolver matures March 2030 | Sources include https://sec.gov/Archives/edgar/data/1616707/000161670726000198/w-20260930.htm, https://spglobal.com/ratings/en/regulatory/article/-/view/type/HTML/id/3609570 and https://investing.com/news/stock-market-news/fitch-upgrades-wayfair-rating-on-profitability-gains-93CH-4685415 The snapshot also carries profitability, ownership, a risk call and a verdict with its reasons. The sections each return a summary, dated findings with a source URL per claim, flagged items and the fields the agent couldn't source. --- # Deal Sourcing Agent > Describe your investment thesis or name a company you like. Nimble agents search startup lists, funding news and hiring signals on the live web and return a ranked deal list with stage, investors, founders and a source for every value. Interactive demo: https://www.nimbleway.com/demos/deal-sourcing ## Find companies that match your thesis before anyone else Describe your thesis or name a company you like, and Nimble agents search startup lists, funding news and hiring signals on the live web for companies that fit. ## Examples to try - "Bootstrapped European B2B software, 20 to 200 employees" - Ramp, Rippling, Deel ## Set it up in your agent The Nimble skill for this use case: https://docs.nimbleway.com/skills/deal-sourcing/SKILL.md Prompt to paste into your agent: ```text Be my deal sourcing assistant using Nimble. Follow https://docs.nimbleway.com/skills/deal-sourcing/SKILL.md. ``` Get an API key: https://online.nimbleway.com/signup ## The most accurate & efficient web search for deal sourcing We asked Opus 5.5 to source companies for investment theses and to answer the questions an investor checks first: stage, ownership, last round, investors and founders. It ran once with each web search provider, and every answer was checked against a hand-verified ground truth. We tracked accuracy and the total cost of getting there, including search fees and tokens. Benchmark code and data: https://github.com/Nimbleway/deal-sourcing-benchmark ## What you get A ranked list of private companies that fit an investment thesis, researched on the live web. Each company carries its website, a one-liner, HQ, founded year, employees, stage and last round, investors, ownership, founders, a recent momentum signal, a fit score from 0 to 100, why it fits and why now. Stage, last round, investors, momentum and founders each link to a source. Stage and ownership show only when a source states them; otherwise they stay unknown, and a company that misses a hard criterion is left out instead of kept to fill the list. A short list is an honest answer. ## Modes | Mode | Input | What it does | |---|---|---| | A thesis | A sentence describing sector, geography, size and ownership | Finds companies that fit the thesis | | Lookalikes of a company | A company name or website, for example Ramp | Finds earlier-stage companies like that one | The demo then narrows by stage (Bootstrapped & founder-owned, Pre-seed–Seed, Series A–B, Growth, PE-backed), geography (Anywhere, United States, North America, Europe, UK & Ireland, Israel, Asia-Pacific) and list size (10 or 20 companies). It searches three angles in parallel: startup lists and directories, funding and deal news, and momentum (hiring, launches, awards). The list exports as CSV, one row per company with name and domain first, or as JSON. ## Inputs | Input | Required | Example | |---|---|---| | Thesis | Yes | Sector, geography, size and ownership in one sentence, for example "Bootstrapped European B2B software, 20 to 200 employees" | | Exclusions | No | For example "no PE-backed" or "not in our portfolio" | ## Run it yourself The Nimble skill for this use case runs the Nimble agent template `deal-sourcing`: https://docs.nimbleway.com/skills/deal-sourcing/SKILL.md - Create the agent once: `nimble_agents_create(template="deal-sourcing", agent_name="deal-sourcing")` - Run it: `nimble_agents_run(agent_name="deal-sourcing", input="Find companies in with . Exclude .", effort="medium")` - Example input: `Find bootstrapped European vertical SaaS companies for logistics or freight forwarding with 20-200 employees` - Effort: the skill runs the template at medium effort (about 1-3 minutes), even when the template's own default is different. Ask for low, medium, high or x-high to run another. State each criterion as a clear test, ownership most of all. For a big market map, split it by region or sub-sector, one run per slice in parallel, then merge and dedupe. Don't ask for a number of companies: narrow theses return few rows on purpose. Ownership is the hardest field, so spot-check the top rows before outreach. ## When to use it, and when not to Use it for "find acquisition targets", "deal sourcing", "map the market for", "founder-owned companies in", "off-market targets" or "companies that fit our thesis" requests. - For a sales prospect list, use Account Discovery: https://www.nimbleway.com/demos/account-discovery.md - To research one company in depth, use Due Diligence: https://www.nimbleway.com/demos/due-diligence.md ## Sample result From the demo's recorded run on the thesis "Vertical SaaS for construction in the US, Seed–Series A", thesis mode, stages Pre-seed–Seed and Series A–B, United States, 10 companies, recorded 2026-11-09. The recording does not state an effort level. The three angle agents found 11 matching companies; the top three: | Rank | Company | Stage & last round | Investors | HQ | Fit | |---|---|---|---|---|---| | 1 | Agave (useagave.com): AI agents that automate construction back-office financial work. | Series A, $15M, 2026-07 | Accel, Y Combinator | San Francisco, CA, United States | 92 | | 2 | Cascade (usecascade.ai): AI-powered business platform for AEC firms. | Seed, $3.5M, 2026-07 | Andreessen Horowitz Speedrun, Ada Ventures, Blitzscaling Ventures, Indico Capital Partners, Shucker VC, Snowball VC, G2C Ventures | New York, United States | 92 | | 3 | Buildcheck (buildcheck.ai): Computer-vision AI that reviews construction drawings for errors. | Series A, $12M, September 2026 | Telescope Partners, WND Ventures, Uncork Capital, Salt VC, Xfund | San Francisco, United States | 90 | Sources include https://useagave.com/blog/announcing-agaves-15m-series-a, https://techcrunch.com/2026/07/22/cascade-raises-3-5m-to-help-construction-firms-find-and-win-projects/ and https://www.fundraiseinsider.com/blog/buildcheck-raises-12m-series-a-for-ai-construction-drawing-reviews/ Each row also carries founders with a short background, why it fits, why now, and which angles found it. Fields no source stated, such as Buildcheck's founded year and employee count, come back empty and show as not published. --- # Decision-maker identification > Find the current buying committee at any account: the economic buyer, champions, technical evaluators, users and gatekeepers for what you sell, each with title, start date, profile and a dated public source. Interactive demo: https://www.nimbleway.com/demos/decision-maker-identification ## Find the right people to talk to at any account Give a company and what you sell, and Nimble agents search the live web for the current buying committee, each person confirmed by a dated public page. ## Examples to try - "Who are the decision makers for sales tools at Datadog?" - Ramp, Datadog, Canva, Shopify ## Set it up in your agent The Nimble skill for this use case: https://docs.nimbleway.com/skills/decision-maker-identification/SKILL.md Prompt to paste into your agent: ```text Be my contact research assistant using Nimble. Follow https://docs.nimbleway.com/skills/decision-maker-identification/SKILL.md. ``` Get an API key: https://online.nimbleway.com/signup ## The most accurate & efficient web search for GTM data We asked Opus 5.5 hundreds of GTM questions about companies, people and industry news. It ran once with each web search provider, and every answer was checked against a hand-verified ground truth. We tracked accuracy and the total cost of getting there, including search fees and tokens. Benchmark code and data: https://github.com/Nimbleway/gtm-enrichment-benchmark ## What you get The people who hold given roles at a company today, researched on the live web. For each role: the person's name, exact title, start date, previous roles, public profiles and a dated source that confirms they are in the role now. Roles the agent can't confirm are listed as missing instead of guessed. It never returns emails or phone numbers. The demo maps a whole buying committee. You pick what you sell, and it runs one agent per part of the deal in parallel. Each person carries a title, seniority, department, start date where found, location, why they matter, a profile link and the source page with its date. People found by more than one search are merged, and people without a dated source are set aside. ## Buying committee | Group | Who it finds | |---|---| | Economic buyer | Owns the budget and signs | | Champions | Lead the team that feels the problem | | Technical evaluators | Test, integrate and run it | | Users | Use it every day or manage those who do | | Gatekeepers | Security, procurement, legal and IT review | The technical evaluators search also returns the users. ## What you sell The product you pick sets the roles the champions search looks for. Seniority can be narrowed to C-level, VP, Director and Manager (all on by default). | Product | Roles | |---|---| | Sales tech | CRO, VP Sales, Head of Revenue Operations, VP Sales Development, Head of Sales Enablement | | Marketing tech | CMO, VP Marketing, VP Demand Generation, Head of Growth, Head of Marketing Operations | | Security | CISO, VP Security, Head of Security Engineering, Head of GRC, Director of Security Operations | | Data & AI | Chief Data Officer, VP Data, Head of Data Engineering, Head of AI or ML, Head of Analytics | | Finance & spend | CFO, VP Finance, Controller, Head of FP&A, Head of Procurement | | HR & people | Chief People Officer, VP People, Head of Talent Acquisition, Head of People Operations, Head of Total Rewards | | IT & infrastructure | CIO, VP IT, Head of Infrastructure, Head of IT Operations, Director of Enterprise Applications | | Developer tools | CTO, VP Engineering, Head of Platform Engineering, Head of Developer Experience, Director of Engineering | | Custom roles | The roles you type | ## Inputs | Input | Required | Example | |---|---|---| | Company | Yes | Name and domain, or a list of them | | Roles | Yes | Up to about 5, for example "CFO, head of RevOps, VP engineering" | ## Run it yourself The Nimble skill for this use case runs the Nimble agent template `decision-maker-identification`: https://docs.nimbleway.com/skills/decision-maker-identification/SKILL.md - Create the agent once: `nimble_agents_create(template="decision-maker-identification", agent_name="decision-maker-identification")` - Run it: `nimble_agents_run(agent_name="decision-maker-identification", input="At (), find the current , , in ", effort="medium")` - Effort: the skill runs the template at medium effort (about 1-3 minutes), even when the template's own default is different. Ask for low, medium, high or x-high to run another. One company per run. For a list, run one per company in parallel. Use the title the company would use: "Head of RevOps" finds more than "Revenue Operations Leader". A role in `missing_roles` means no current source confirmed it. LinkedIn profile pages need a login, so the agent relies on company pages, press and public posts. If the company has no such role, it returns the closest one and explains why. ## When to use it, and when not to Use it for "who is the CFO at", "find the head of X at", "find the buying committee", "decision makers at", "who runs RevOps at" or "find contacts at" requests. - For company-level fields such as headcount, funding or HQ, use Account Enrichment: https://www.nimbleway.com/demos/account-enrichment.md - For a brief before the call, use Account Brief: https://www.nimbleway.com/demos/account-brief.md - For what changed at an account recently, use Account Signals: https://www.nimbleway.com/demos/account-signals.md ## Sample result From the demo's recorded run on Ramp (ramp.com), Sales tech, all seniority levels, four committee searches at low effort, recorded 2026-11-09. Each search graded its result medium confidence. Titles as recorded: | Group | Titles found | Roles not confirmed | |---|---|---| | Economic buyer | SVP of Sales; Co-Founder & Co-CEO | Chief Revenue Officer | | Champions | SVP of Sales; VP of Sales; Director GTM Finance & Operations | Chief Revenue Officer, Head of Revenue Operations, VP Sales Development, Head of Sales Enablement | | Technical evaluators and users | Senior GTM Systems Analyst; Senior GTM Systems Engineer; Manager, Sales Development (2); Commercial Sales Manager | Salesforce/CRM Administrator, Revenue/Sales Operations Manager | | Gatekeepers | General Counsel; Head of Ramp Procurement; Head of Business Systems Operations; Head of Infrastructure & DevOps | Chief Information Security Officer (CISO), Head of Security | Source: https://exa.ai/websets/directory/ramp-executives --- # Due Diligence Agent > Run deep due diligence on any company. Financials, leadership, litigation and reputation from the live web, with a source, date and confidence grade on every finding. Interactive demo: https://www.nimbleway.com/demos/due-diligence ## Run deep due diligence on any company Research a specific company with Nimble agents that search deep across the live web, reading company sites, filings, court and regulator records, news and reviews. ## Examples to try - "Run due diligence on Rippling" - Rippling, Deel, Ramp, Canva, Nvidia ## Set it up in your agent The Nimble skill for this use case: https://docs.nimbleway.com/skills/due-diligence/SKILL.md Prompt to paste into your agent: ```text Be my due diligence assistant using Nimble. Follow https://docs.nimbleway.com/skills/due-diligence/SKILL.md. ``` Get an API key: https://online.nimbleway.com/signup ## The most accurate & efficient web search for due diligence We asked Opus 5.5 hard diligence questions about companies: funding, leadership, litigation, financials and reputation. It ran once with each web search provider, and every answer was checked against a hand-verified ground truth. We tracked accuracy and the total cost of getting there, including search fees and tokens. Benchmark code and data: https://github.com/Nimbleway/due-diligence-benchmark ## What you get A first-pass due diligence memo on one company, researched on the live web. It covers the business, traction, team, funding, financial signals, legal and regulatory issues, reputation and red flags, and ends with a short verdict. Every claim has a date and a source, each finding carries a confidence grade, and fields the agent couldn't source are listed as missing instead of guessed. ## Modes The demo opens with a snapshot of the company, then researches the sections for the role you pick. | Mode | For | Sections | Snapshot | |---|---|---|---| | Venture investment | Private companies | Traction, Team & founders, Funding & cap table, Market & competitors, Red flags | ARR / revenue, growth, last round, valuation, lead investors, headcount, plus a bull case, a bear case and questions for the founders | | Public equity research | Listed companies | Catalyst calendar, Results vs guidance, Insider & ownership moves, Legal & regulatory, News flow | Ticker, market cap, last quarter, guidance, next earnings, key catalyst, key risk | | Private equity & M&A | Acquisition targets | Business quality, Customers & concentration, Management depth, Liabilities, Ownership & deal history | Revenue, profitability, growth, customer concentration, ownership, employees | | Counterparty & compliance | Onboarding and KYB | Registry & entities, Officers & owners, Sanctions & PEP, Adverse media | Legal name, jurisdiction, registration, LEI | | General due diligence | Any company | Financials, Leadership, Litigation, Reputation, Red flags, News, Market & competitors | Founded, headquarters, CEO, employees, total funding, latest valuation, ownership | ## Inputs | Input | Required | Example | |---|---|---| | Company | Yes | Name and domain, for example Canva (canva.com) | | Focus | No | Anything to check first, for example "focus on the AI cost story" | ## Run it yourself The Nimble skill for this use case runs the Nimble agent template `due-diligence`: https://docs.nimbleway.com/skills/due-diligence/SKILL.md - Create the agent once: `nimble_agents_create(template="due-diligence", agent_name="due-diligence")` - Run it: `nimble_agents_run(agent_name="due-diligence", input="Write a due diligence memo on (). Focus on .", effort="medium")` - Effort: the skill runs the template at medium effort (about 1-3 minutes), even when the template's own default is different. Ask for low, medium, high or x-high to run another. One company per run. A focus line shifts depth, it does not remove sections. ## When to use it, and when not to Use it for "due diligence on", "DD memo", "investment memo on", "red flags for" or "should we invest in" requests about a company. - For a brief before a sales meeting, use Account Brief: https://www.nimbleway.com/demos/account-brief.md - For whether a company can pay, use Credit Risk: https://www.nimbleway.com/demos/credit-risk.md - For sanctions, PEP and adverse-media screening of a counterparty, see KYC Screening: https://www.nimbleway.com/demos/kyc-screening.md ## Sample result From the demo's recorded run on Rippling (rippling.com), Venture investment mode, company snapshot, low effort, recorded 2026-11-09. The agent ran 6 searches and 27 extracts and graded the snapshot high confidence. | Field | Value | |---|---| | Revenue | ~$1B ARR as of Q1/Q2 2026 — company-stated (CEO Parker Conrad on X, per GetLatka) and press-confirmed; up from ~$850M at end-2025 | | Growth | ~78% YoY ARR growth as of March 2026 (press-reported; consistent with the company-stated $1B ARR milestone) | | Last round | Series G, $450M, May 2025, at a $16.8B valuation | | Valuation | $16.8B at the May 2025 Series G (priced round); secondary-market implied value ~$19.0B as of Nov 8, 2026 (StockAnalysis/Clarity estimate) | | Headcount | 5,000 employees as of November 2025 (GetLatka), up from 3,630 in September 2024; a separate aggregator index estimates ~7,900 (undated) | Sources include https://getlatka.com/companies/rippling, https://stockanalysis.com/private/rippling/valuation and https://news.outsourceaccelerator.com/rippling-raises-series-g-funding The other sections (traction, team, funding, market, red flags) each return a summary and a list of dated findings with a source URL and a confidence grade per claim. --- # Job Hunt > Find live roles that fit you, who owns each hire and what they said lately, and a short email to each, from the live web. Pick what you optimize for, see Nimble's plan, and run it in your own agent. Interactive demo: https://www.nimbleway.com/demos/job-hunt ## Find the job, then the people who hire for it Tell Nimble what you're after. It plans the hunt on the live web: open roles from the last 30 days ranked by fit, the hiring manager and the executive above them at each, what they said lately, and a short email to each. Your own agent runs the plan. ## Examples to try - "Find senior product manager jobs in New York" ## Set it up in your agent Get an API key: https://online.nimbleway.com/signup ## What you get Nimble's plan for a job hunt, run by your own agent: live postings from the last 30 days that fit your target titles, seniority, skills and location, ranked by fit (company, role, location, date posted, link and why you fit); at each company, the hiring manager who runs the team and the executive above them (a founder at a small company), with how sure the plan is that each owns the hire; what each said publicly lately (LinkedIn and X posts, talks, podcasts, interviews and blog posts, each dated and linked, with company news labelled as such); and a short email to each person that ties your experience to something they said. It finds no email addresses or phone numbers. ## Inputs | Input | Required | Example | |---|---|---| | Job titles | Yes, up to 3 | Senior Product Manager, Group Product Manager | | Where | No | New York, or Remote | | Country | No | US | | Key skills | No | B2B SaaS, payments, API products | Your agent asks for your resume before it ranks the roles; it stays in your chat. ## Optimize for | Choice | Plan | |---|---| | Fast and cheap | Job boards and LinkedIn hiring posts, the best postings read and dated, one people search for who leads the hire. No research run (about 2 minutes) | | Balanced | Job boards and hiring posts, every posting read and dated, one research row per job for the people and what they said lately (5 to 9 minutes) | | Thorough | The widest search of job boards and hiring posts, one research row per job (5 to 9 minutes) | ## Run it yourself Connect Nimble's Web Gateway MCP server (`https://mcp.webrouter.net/mcp`) to your agent, then call `web_gateway_plan` with `{"preset": "nimble.jobs.hunt", "input": [{"title": "Senior Product Manager", "location": "New York", "country": "US"}]}` and run the steps it returns. --- # KYC Screening > Screen any business for sanctions, PEPs and adverse media from the live web. Registry identity, owners and officers, sanctions and watchlists, PEP exposure, adverse media and licensing, every finding with a source and date, as a case file. Interactive demo: https://www.nimbleway.com/demos/kyc-screening ## Screen any business for sanctions, PEPs and adverse media Give a business name or website. Nimble agents check registries, owners, sanctions lists, PEP exposure, adverse media and licences on the live web and return a sourced case file. ## Examples to try - "Screen Binance for sanctions and adverse media" - Stripe, Wirecard, Binance, Garantex ## Set it up in your agent The Nimble skill for this use case: https://docs.nimbleway.com/skills/kyc-screening/SKILL.md Prompt to paste into your agent: ```text Be my KYC research assistant using Nimble. Follow https://docs.nimbleway.com/skills/kyc-screening/SKILL.md. ``` Get an API key: https://online.nimbleway.com/signup ## The most accurate & efficient web search for compliance research We asked Opus 5.5 screening questions about businesses: registry details, owners and officers, sanctions designations, enforcement actions and adverse media. It ran once with each web search provider, and every answer was checked against a hand-verified ground truth. We tracked accuracy and the total cost of getting there, including search fees and tokens. Benchmark code and data: https://github.com/Nimbleway/kyc-screening-benchmark ## What you get A sourced KYC / KYB case file on one business, researched on the live web: registry details, officers and beneficial owners, sanctions and PEP checks, adverse media, licences and enforcement, related entities, the namesakes it excluded, and a risk rating with its drivers. Every finding has a source URL and a date, and fields the agent couldn't find are listed as missing. It is a research screen from public web sources, not a certified sanctions or PEP list check. ## Checks The demo screens businesses only. Owners and officers found in public registries are screened in their roles at the business. Each check returns a status of Clear, Review or Hit, a summary and dated, sourced findings. | Check | What it covers | |---|---| | Registry & identity | Legal entity, registration number, status and registered address | | Ownership & control | Directors, officers and beneficial owners | | Sanctions & watchlists | OFAC, UK, EU and UN lists for the business and its people | | PEP exposure | Politically exposed owners and officers, and state control | | Adverse media | Fraud, money laundering, bribery, sanctions evasion and more | | Regulatory & licensing | Licences held and enforcement actions | The entity card shows legal name, registration number, jurisdiction, status, incorporation date, registered address, website and LEI, why it is the same entity, and related entities. The risk rating is computed from the check statuses: any hit makes it High, any review (or a check that did not finish) makes it Medium, and all clear is Low. The case file downloads as PDF or copies as JSON. ## Inputs | Input | Required | Example | |---|---|---| | Company | Yes | Legal or trading name | | Country | Yes | Where it is registered | | Registry number or LEI | No | Removes namesake risk | ## Run it yourself The Nimble skill for this use case runs the Nimble agent template `kyc-screening`: https://docs.nimbleway.com/skills/kyc-screening/SKILL.md - Create the agent once: `nimble_agents_create(template="kyc-screening", agent_name="kyc-screening")` - Run it: `nimble_agents_run(agent_name="kyc-screening", input="Run a KYC screen on in . Registry number: . Website: .", effort="medium")` - Effort: the skill runs the template at medium effort (about 1-3 minutes), even when the template's own default is different. Ask for low, medium, high or x-high to run another. One entity per run, and country is required. A registry number or LEI makes the match exact. One trading name can map to several legal entities, so check the registration block is the one you meant. Coverage is strongest where public registries exist: the UK, the EU and US states. ## When to use it, and when not to Use it for "KYC on", "KYB", "AML screen", "sanctions check", "beneficial owners of", "screen this company" or "onboarding check" requests about a business. - For whether a business can pay, use Credit Risk: https://www.nimbleway.com/demos/credit-risk.md - For a full due diligence memo on a company, use Due Diligence: https://www.nimbleway.com/demos/due-diligence.md ## Sample result From the demo's recorded Registry & identity check on Stripe (stripe.com), United States, no registration number given, recorded 2026-11-09. The recording does not state an effort level. The agent ran 6 searches and 29 extracts and returned the check as Clear. | Field | Value | |---|---| | Legal name | STRIPE, LLC | | Registration number | 4675506 | | Jurisdiction | Delaware, United States (US-DE) | | Status | ACTIVE | | Incorporated | 2009-04-13 | | LEI | 549300CLHGIPTCYHQ143 | Sources include https://lei.bloomberg.com/leis/view/549300CLHGIPTCYHQ143, https://lei.bloomberg.com/leis/view/254900VYCQ4K0SLSXF22 and https://opencorpdata.com/lei/549300CLHGIPTCYHQ143 The check also listed two related group entities, the parent Stripe Global Holdings Inc. and the affiliate Stripe Payments Company, and excluded 2 namesakes. The other checks (ownership, sanctions, PEP, adverse media, regulatory) each return a status, a summary and dated findings with a source URL. --- # Portfolio Monitoring > Know what changed across your portfolio. Nimble agents check the live web for funding, leadership, layoffs, lawsuits, distress and deals at each company you hold, and return a dated, sourced digest, worst first. Interactive demo: https://www.nimbleway.com/demos/portfolio-monitoring ## Know what changed across your portfolio this week Add the companies you hold. Nimble agents check the live web for what changed at each one and list it worst first, dated and sourced. ## Examples to try - "What changed at my portfolio companies this month?" ## Set it up in your agent The Nimble skill for this use case: https://docs.nimbleway.com/skills/portfolio-monitoring/SKILL.md Prompt to paste into your agent: ```text Be my portfolio monitoring assistant using Nimble. Follow https://docs.nimbleway.com/skills/portfolio-monitoring/SKILL.md. ``` Get an API key: https://online.nimbleway.com/signup ## The most accurate & efficient web search for portfolio monitoring We asked Opus 5.5 what changed at hundreds of private and public companies: funding, leadership, layoffs, lawsuits, deals and distress. It ran once with each web search provider, and every answer was checked against a hand-verified ground truth. We tracked accuracy and the total cost of getting there, including search fees and tokens. Benchmark code and data: https://github.com/Nimbleway/portfolio-monitoring-benchmark ## What you get A digest of what changed at each company you hold, researched on the live web. Each company gets a status (Needs attention, Watch or Steady) and one line on what changed. Each event carries the date it happened (not the article date), a category, a severity (high, medium or low) and a source URL. Only events inside the window count: re-reported old news and press-release filler are left out, and the digest lists what it left out and why. If nothing material happened, the company comes back Steady with an empty list. ## What it watches | Group | Topics | |---|---| | Money | Funding & valuation, Financial distress, M&A | | People | Leadership, Hiring & layoffs | | Business | Product & partnerships, Customers & contracts | | Risk | Legal & regulatory, Reputation | All topics except Reputation are on by default. The window is the last 7, 30 or 90 days. The demo checks each company with its own agent, all in parallel, and sorts the digest worst first. It exports as CSV and can draft an LP update or a board note from the results. ## Inputs | Input | Required | Example | |---|---|---| | Holdings | Yes | Names or tickers, or a list | | Since | No | A date in YYYY-MM-DD. Defaults to 30 days ago | | Materiality rule | No | For example "ignore product launches". Defaults to the template's rule | ## Run it yourself The Nimble skill for this use case runs the Nimble agent template `portfolio-monitoring`: https://docs.nimbleway.com/skills/portfolio-monitoring/SKILL.md - Create the agent once: `nimble_agents_create(template="portfolio-monitoring", agent_name="portfolio-monitoring")` - Run it: `nimble_agents_run(agent_name="portfolio-monitoring", input="Check () for material events since ", effort="medium")` - Effort: the skill runs the template at medium effort (about 1-3 minutes), even when the template's own default is different. Ask for low, medium, high or x-high to run another. One holding per run. For a portfolio, run one per holding in parallel. To apply your own materiality rule, add one line to the input. Keep only events strictly after the since date, since events on the date itself can slip in. To keep watching, schedule the run and pass the last run date as the new since date. ## When to use it, and when not to Use it for "what happened at [holding]", "monitor my portfolio", "material events", "check my holdings since" or "portfolio news" requests. - For changes at sales accounts, use Account Signals: https://www.nimbleway.com/demos/account-signals.md - For a full credit view of one company, use Credit Risk: https://www.nimbleway.com/demos/credit-risk.md - For a deep look at one holding, use Due Diligence: https://www.nimbleway.com/demos/due-diligence.md ## Sample result From the demo's recorded run on Mercury (mercury.com), all default topics, a 30-day window since 2026-10-11, high effort, recorded 2026-11-10. Status: Watch. Summary as recorded: "Two product launches — Mercury Payroll (Oct 20) and AI-agent money movement via MCP (Oct 15); no funding, leadership or legal changes." | Date | Category | Severity | Event | |---|---|---|---| | 2026-10-20 | Product & partnerships | medium | Mercury launched Mercury Payroll, an AI-native payroll product handling tax and compliance in all 50 states, free through Dec 31, 2026 then $100/month plus $50/month per person paid. | | 2026-10-15 | Product & partnerships | medium | Mercury launched actions on its MCP, letting AI agents like ChatGPT and Claude queue money moves (transfers, vendor payments) that complete only after human approval. | Sources include https://www.businesswire.com/news/home/20261020220473/en/Mercury-Launches-AI-Native-Payroll and https://finance.yahoo.com/small-business/articles/mercury-launches-ai-native-payroll-150000778.html The run also left out 5 items, each with a reason: news dated before the window, and analysis re-reporting an earlier round rather than a new event. --- # Shopping Search > Find the products reviewers recommend that fit your budget and needs, each on a store page with today's price and stock, from the live web. Pick what you optimize for, see Nimble's plan, and run it in your own agent. Interactive demo: https://www.nimbleway.com/demos/shopping-search ## Find what to buy, at today's price Tell Nimble what you want, budget and all. It plans the search on the live web: the products reviewers recommend that fit every constraint, each on a store page with the price that page shows today and whether it is in stock. Your own agent runs the plan. ## Examples to try - "Wireless headphones under $100" ## Set it up in your agent Get an API key: https://online.nimbleway.com/signup ## What you get Nimble's plan for a shopping search, run by your own agent: the products independent reviewers recommend that fit every constraint in your request (type, budget, size, compatibility, features), each on a store page (the brand's own store, a retailer or a marketplace listing) with the price that page shows today, its currency, whether it is in stock, why it fits and the reviews that recommend it. A product over your budget is dropped, colour variants count as one product, and anything the plan could not find or price is listed as a gap. Your agent buys nothing. ## Inputs | Input | Required | Example | |---|---|---| | What you are shopping for | Yes | Wireless headphones under $100 | Name a country in the request ("in the UK") for that country's stores and currency; US dollars otherwise. ## Optimize for | Choice | Plan | |---|---| | Fast and cheap | A review search and a store search, the best store pages read (about 3 minutes) | | Balanced | A shortlist of 4 recommended products, a price search for each, up to 12 store pages read (2 to 4 minutes) | | Thorough | Owner threads on Reddit too, up to 6 products, up to 20 store pages read (4 to 5 minutes) | ## Run it yourself Connect Nimble's Web Gateway MCP server (`https://mcp.webrouter.net/mcp`) to your agent, then call `web_gateway_plan` with `{"preset": "nimble.search.products", "query": "Wireless headphones under $100"}` and run the steps it returns.