1. Connect the plugin once
Add the Alpha Sophia marketplace and install the plugin in Claude. Sign in with your Alpha Sophia account when prompted — OAuth, no API key. Claude is now connected to your licensed data backbone.
Connect Claude to Alpha Sophia — the leading commercial healthcare intelligence platform — and answer targeting, KOL, and market-sizing questions on real, current, sourced US provider data. No more hallucinated names and numbers.
Official Alpha Sophia plugin for Claude
1,976 Texas cardiologists actively managing heart failure (last 12 months). Top targets by HF patient volume:
Ranked on current, real-world data — sourced and safe to act on.
Illustrative example of a plugin-powered answer in Claude.
Prefer ChatGPT or Cursor, or want fully autonomous workflows? See all the ways to connect AI agents to Alpha Sophia over MCP.
A general-purpose model can't see your market. The plugin fixes the gaps that make raw AI risky for commercial and medical decisions.
Models have a training cutoff — they can't see current procedure volumes, prescribing, affiliations, or who's publishing now.
On their own, models invent provider names and numbers you have no way to verify — too risky to build strategy on.
Claims, NPI, and payments data are licensed and siloed — not on the open web a general model was trained on.
A wrong target list wastes scarce rep time and budget. Decisions need defensible, sourced numbers.
Trials, practice moves, and KOL influence change every quarter — last year's snapshot is already stale.
HIPAA, the Sunshine Act, and MLR review demand outputs you can audit and defend — not a black box.
The fix: a vetted data plugin that gives Claude real, current, sourced data.
Same question. The plugin is the difference between a hedged guess and a ranked, sourced answer you can act on.
"Which cardiologists should my reps prioritize for our heart-failure launch in Texas?"
"I can't give a reliable name-level list — I have no live provider data and a training cutoff. Directionally, look at the big advanced-heart-failure centers, then layer in your own claims/CRM data to rank individuals."
1,976 Texas cardiologists actively managing heart failure (last 12 months). Top targets by HF patient volume, ranked and affiliation-linked.
"Who are the top heart-failure KOLs we should engage before launch?"
"I can name prominent HF experts, but treat them as starting points, not a ranking. Affiliations change and my data has a cutoff, so I can't rank by recent publications or active trials."
163,733 HF publications in the last 12 months · 712 active trials. Leading KOLs ranked by current publications and trial activity, each sourced and affiliation-linked.
Illustrative examples based on Alpha Sophia data.
The plugin gives Claude both the tools (over MCP) and the playbook, so it queries Alpha Sophia the same disciplined way our in-app assistant does.
Add the Alpha Sophia marketplace and install the plugin in Claude. Sign in with your Alpha Sophia account when prompted — OAuth, no API key. Claude is now connected to your licensed data backbone.
No filter-building, no query language. Ask the way you'd ask a colleague — "how many", "find and rank", "how big is the market" — and Claude figures out which Alpha Sophia tools to call.
Behind the scenes, Claude resolves your terms (a specialty, a drug, a condition, a city) to canonical IDs, builds a structured filter, and runs it against Alpha Sophia — the same parse, build, execute pipeline as the app. Numbers come from data, not memory.
A ranked target list, a market-size number by state, a KOL shortlist — with the filters Claude applied stated in plain English so you can trust and refine the result, then take it into Alpha Sophia to export or sync to your CRM.
The same data backbone powers every role — so each team moves from days of research to minutes of answers.
Needs: HCP volumes, affiliations & contacts
Needs: Geographic & procedure trends
Needs: Sites of care, by geography
Needs: Publications, trials & KOLs
Needs: Clean, exportable target lists
Needs: Market sizing & benchmarks
A single answer is great. A finished deliverable is better. In Claude Cowork, Claude works as an agent — chaining multiple Alpha Sophia calls and its own reasoning into a complete piece of work.
A chatbot is plenty for a quick answer or a reworded email. For a ranked, sourced, exportable deliverable, use an agent.
"Build and rank a Texas heart-failure target list from live data, then draft a one-line territory note for each rep."
"Create a pre-launch KOL map: top authors and investigators for our indication, with affiliations and recent output."
"Produce a market-sizing brief by state for our therapy, with sites of care and procedure volumes, laid out as a table."
Every Alpha Sophia entity is available to the plugin — the same data that powers the web app, scoped to your organization.
Two commands in Claude Code, then sign in. Your organization needs an Alpha Sophia subscription with the API and Agent modules enabled.
# 1. Add the Alpha Sophia marketplace
/plugin marketplace add alphasophia/claude-plugins
# 2. Install the plugin
/plugin install alpha-sophia@alpha-sophia
# 3. Reload, then sign in when Claude first calls a tool
/reload-pluginsNo API key. Claude prompts you to sign in with your Alpha Sophia account on the first question.
Every answer is drawn from your licensed data and metered against your agent token balance.
The plugin uses the same pipeline as the Alpha Sophia web app, so answers stay consistent.
Not sure if your plan includes plugin access? Talk to our team.
Same data backbone — every role moves from days of research to minutes of answers.
Highest-volume targets for the quarter, with affiliations and contacts.
Which metros over-index for a procedure — so you know where to spend.
How many sites of care perform a procedure, and exactly where they are.
Who's publishing and running trials on your topic right now.
Clean, exportable target lists with volumes and affiliations.
Addressable market sizing and benchmarks by therapy and geography.
Access is authenticated per user and scoped to your organization, under the same enterprise-grade controls trusted by 100+ life science customers. Because answers are sourced from Alpha Sophia rather than the model's memory, they're traceable and auditable. As with any AI in a regulated setting, keep a human in the loop before outputs inform irreversible actions or regulated content.
It's an official plugin that connects Claude to Alpha Sophia — the leading commercial healthcare intelligence platform. Once installed, Claude can answer targeting and insights questions using real, current, sourced US provider data instead of guessing from its training data. It bundles our MCP (Model Context Protocol) server and a built-in playbook so Claude uses the data the same way our in-app assistant does.
The plugin connects Claude to the same data backbone that powers the Alpha Sophia web app, scoped to your organization. You sign in with your Alpha Sophia account, and every answer is drawn from your licensed data — providers (HCPs), organizations (HCOs), sites of care, clinical trials, investigators, publications, and authors. Your org needs the API and Agent modules enabled, an active subscription, and available agent tokens.
MCP (Model Context Protocol) is the open standard Claude uses to call external tools. The Alpha Sophia plugin exposes tools like resolve_entity, search/count/market_size for providers and sites of care, and search/count for clinical trials, investigators, publications, and authors. When you ask a question, Claude resolves your terms to canonical IDs, builds a structured filter, and runs it against Alpha Sophia — no raw SQL, no hallucinated numbers.
In Claude Code, add our marketplace and install the plugin: "/plugin marketplace add alphasophia/claude-plugins" then "/plugin install alpha-sophia@alpha-sophia". Restart or run "/reload-plugins", and sign in with your Alpha Sophia account when prompted on the first question.
No. The plugin uses OAuth: the first time Claude calls an Alpha Sophia tool, it prompts you to sign in with your Alpha Sophia account and approve access. There is no API key to copy or environment variable to set.
Anything you would use Alpha Sophia for: building and ranking target lists, market sizing, finding sites of care, mapping KOLs by publications and trials, and geographic or procedure trend analysis. Ask in plain English — for example, "Which cardiologists should my reps prioritize for our heart-failure launch in Texas?" or "Size the addressable market for this therapy by state."
Claude Cowork lets Claude work as an agent that chains multiple steps into a finished deliverable. With the Alpha Sophia plugin connected, Claude can resolve a disease and geography, pull and rank providers by volume, cross-reference KOLs by publications and trials, and assemble an exportable target list or territory plan end to end — turning days of research into minutes.
Yes. Because every answer is generated from Alpha Sophia data rather than the model’s memory, results are current and traceable to the underlying provider, site, trial, publication, and author records — so they are defensible enough to build commercial strategy on.
Commercial and medical teams across life sciences and healthcare: field sales and targeting, marketing, market access, medical affairs and MSLs, sales operations, and commercial insights. Everyone works from the same data backbone, so answers stay consistent across roles.
Access is authenticated per user and scoped to your organization, under the same enterprise-grade controls trusted by 100+ life science customers. As with any AI workflow in a regulated environment, review outputs before they inform irreversible actions or regulated content.
Make informed connections with the most suitable healthcare providers for your product.