Mamba Labs

Resolves a company domain, or a batch list of domains in one run, to LinkedIn company URLs, one row per domain, via pattern matching and web search. Every row carries resolution_method, confidence and a slug_mismatch flag, so a weak match is visible rather than silent. Flat, Clay-ready, MCP-ready.

Resolves a company name to its primary domain and LinkedIn URL, one company or a batch of names and domains with one row per entry. Any mix of name, domain, or LinkedIn URL in; one canonical identity out with 0-100 confidence scores, conflict detection, and a match method. Flat, Clay-ready.

Domain to structured company firmographics: employee band, industry, HQ, founded year, revenue estimate, logo, and description from schema.org JSON-LD and meta tags. Flat JSON, Clay ready, with source provenance.

Returns a research grade popularity RANK for any company domain from the public Tranco list, with a 40 day trend and an honest band. Ranks are not visit counts and this actor never invents one. Optional Cloudflare Radar and Chrome UX Report with your own keys. One flat row per company.

Find people at any company by seniority and department. Four-layer cascade: website, SERP, and database enrichment with company-match scoring. Two-provider email waterfall (Icypeas, Prospeo) plus catch-all verification (Reoon, BounceBan). No direct LinkedIn scraping.

Resolves a company domain to its LinkedIn company page and returns the exact follower count plus industry, company size band, headquarters and founded year from the public page. Flat, Clay ready, one row per company.

Keyword-driven job posting monitor that finds companies hiring for any role across job boards. Returns flat, Clay-ready, firmographic-enriched rows with company LinkedIn URL per posting.

Market in, companies out. Give it job role keywords and a location and it returns companies currently hiring for that, built from a live index of public Greenhouse and Ashby job boards. Second mode finds US public companies whose SEC filings mention your phrase. No paid data source.

Finds creators by keyword and niche across TikTok, Instagram, YouTube, Pinterest, Twitch, Threads, and podcasts. Follower band filter keeps unknown counts and skips podcasts. One flat row per creator with handle, profile URL, and follower count. Contributes to a shared creator pool, on by default.

Runs discovery, profile read, and link-in-bio check in one run, from keywords or a handle list. One flat row per creator per platform, linked by a creator ID, with contact, newsletter status, what they sell, and manager. Contributes to a shared creator and agency pool, on by default.

Reads public creator profiles from handles or URLs. One flat row per profile: followers, posts, bio, bio link, email, engagement, verified. A podcast Apple refuses 5 times is labeled source_blocked. max_residential_reads caps the Instagram charge. Contributes to a shared creator pool, on by default.

Follows a creator's link-in-bio page and website. One flat row with links, newsletter status, what they sell, manager contact, evidence URLs, and segment_tier. A file over max_download_mb (25 by default) is skipped as file_too_large. Contributes to a shared creator and agency pool, on by default.

Combines GTM hiring signals and tech stack detection into one flat Clay-ready row with composite score, recommended action, and plain-English summary. Runs two sub-actors in a single call. Built for Clay enrichment and outbound qualification. MCP-ready for Claude Desktop and AI agents.

Company domain in, the executives and staff published on that company's own team, leadership or about page out. Returns names, titles, and page URL, with the pages actually reached so you can tell a real absence from an unreadable site. No LinkedIn scraping.

Audit a domain's email deliverability: SPF, DKIM, DMARC and policy, MX and mail provider, catch-all and blacklist checks, domain age, and a 0 to 100 health score. Flat JSON, Clay ready.

Company domain in, the registered legal entity out: legal name, company number, jurisdiction, status, LEI and VAT number. Only returns a match when the legal names are identical after normalization, so an unresolved domain comes back null instead of a confidently wrong company number.

Can an AI agent read this site. Checks llms.txt, robots AI crawler rules and the Content Signal policy, structured data, render mode, and machine readable endpoints like sitemap, OpenAPI and well known files. Every field is a fact read off a fetch, never a model opinion.

Detects whether a company declares, deploys, or charges for AI. Returns a flat Clay-ready row with an AI maturity tier, detected AI vendors, llms.txt status, and pricing-page evidence.

Measures how much long-form work a company publishes per month and whether that rate is rising or falling. Returns a flat Clay-ready row with 30 day, 90 day and 12 month post counts, a cadence trend, format mix and byline count. Editorial output volume, not changelog monitoring.

Resolves a company domain to its official Bluesky account and returns exact follower, following and post counts through the public AT Protocol API. Flat, Clay ready, one row per company. Counts are exact, not rounded.

Pushes Clay-enriched leads into an Instantly or Smartlead campaign. Maps Clay column names to sequencer fields, gates on ICP score, deduplicates against the campaign, and batches the push. Dry run returns the exact payload without sending. Instantly API v2 and Smartlead API v1.

Detects what changed at a company across hiring, tech stack, funding, firmographics, and social. Returns only the deltas as typed change events in flat, Clay-ready JSON for RevOps and outbound.

Contact title in, department and seniority out, plus a check on whether the person still holds the role at that company. Runs a deterministic title ladder with no API key needed. Bring your own LLM key only for the small share of titles the rules cannot place.

Rechecks a saved creator list and returns one row per change since the last run: follower milestone, newsletter added, new product, new bio link, new manager. A run with no change returns one free no_change row: list name, handle count, run time. Contributes to a shared creator pool, on by default.