OpenFunnel Agent Primitives

@OpenFunnel find 50 companies hiring for React developers in the US, then show me their funding stage and employee count

Returned 50 companies matching React in recent job postings, filtered to USA. Example results: Acme Corp (Series B, 150–300 employees), Widget Labs (Seed, 11–50 employees), DataFlow Inc (Series C, 301–1000 employees). This search consumed 25 credits for the base page; enrichment filters added 10 credits. A next_cursor is available to fetch the next 50 companies if needed.

Tech-stack search is OpenFunnel's core strength—it surfaces companies by what they're actively hiring for, not static firmographic data. Each 50-company page costs credits (25 base + enrichment overhead), so budget accordingly. The MCP requires an API_KEY with sufficient credit balance; check GET_CREDIT_BALANCE first if running large queries.

@OpenFunnel research Widget Labs and tell me what roles they've posted in the last 90 days, plus current team size in engineering

Widget Labs has posted 8 engineering roles in the past 90 days: 3 Frontend Engineers (React), 2 Backend Engineers (Python/Django), 1 DevOps Engineer, 1 Data Scientist, 1 Engineering Manager. Current engineering team estimated at 22 people based on LinkedIn profiles. Evidence cited from job boards (Lever, Greenhouse) and LinkedIn company page as of the query date. This call consumed 1 credit.

Research Company qualifies a single known company with hiring signals and team composition—ideal for account prioritization or competitive intel. Each call costs 1 credit and requires a valid company identifier (domain, LinkedIn URL, or name). The MCP surfaces cited sources, but data freshness depends on OpenFunnel's crawl schedule (typically weekly).

@OpenFunnel start a bulk lookalike search for companies similar to our top 5 customers—Series A to C, 50–500 employees, US-based—then poll until done and summarize the top 20 matches

Bulk lookalike job initiated (job_id: blk_abc123). Status: running. Polling every 30 seconds… Job completed after 4 minutes. Retrieved 200 lookalike companies; top 20 by similarity score: TechCo (0.92), InnovateLabs (0.89), CloudStart (0.87)… All matches fall within Series A–C, 50–500 employees, US location filters. Total cost: 150 credits (5 seed companies × 30 credits/lookalike run).

Bulk lookalike is OpenFunnel's highest-value workflow for prospecting—it finds companies that mirror your ICP across hiring patterns and firmographics. Jobs run asynchronously; the MCP must poll GET_BULK_LOOKALIKE_RESULTS until status reaches 'completed'. Credit cost scales with seed count and result volume. Cancel via CANCEL_BULK_LOOKALIKE_SEARCH if the job runs too long.

When bulk lookalike search beats manual prospecting

A 2-person sales team at a dev-tools startup needs 500 qualified accounts that match their best customers' tech stack. They run a bulk lookalike search filtering for Series A companies using React and Postgres, then poll results with Get Bulk Lookalike Results until the job completes. The MCP handles pagination and credit tracking automatically, so the SDRs export a CSV and load it into their CRM in one morning instead of spending three days scraping LinkedIn. This works when your ICP is technology-driven and you can afford 25+ credits per page. If your filters are purely firmographic (revenue, headcount) or you need real-time enrichment during a call, a traditional data provider is faster. Buy this MCP if you're running weekly list builds and your team already thinks in tech-stack terms.

Real-time company qualification without leaving your notes

A 6-person customer success team at a B2B SaaS company uses Research Company during onboarding calls to surface hiring signals and team composition before the prospect mentions their roadmap. The CSM pastes the company domain into Switchy, the MCP returns cited evidence of recent engineering hires or open roles, and the team tailors the demo on the fly. Each lookup costs credits, so they reserve it for mid-market and enterprise accounts where the context justifies the spend. This breaks down if you're qualifying hundreds of inbound leads daily—credit burn becomes prohibitive and you should pre-enrich in your CRM instead. Use this MCP when discovery calls are high-touch and you need hiring intelligence that's fresher than your enrichment vendor's monthly refresh.

Monitoring competitor tech adoption at scale

A 4-person product marketing team tracks which companies in their vertical adopt a competing platform by running weekly tech-stack searches for the competitor's name. They use Search Companies by Tech Stack to pull 50-company pages, compare results against last week's snapshot in a shared spreadsheet, and flag new adopters for the sales team. The MCP's fixed-page model and credit transparency make budgeting predictable: 25 credits per page means 200 credits for 400 companies monthly. This approach wins when you're monitoring 3-5 specific technologies and can tolerate weekly cadence. If you need daily alerts or want to track 50+ tools simultaneously, a dedicated competitive-intel platform with webhooks is worth the premium. Buy this if your competitive set is narrow and your team already lives in Switchy for other workflows.