# Healthspan Intel Muse connector

From musedirectory.ai, the independent directory of Meta Muse connectors. Not affiliated with Meta.

## Healthspan Intel

What published research says about longevity supplements and protocols, with PubMed citations.

- Record: https://musedirectory.ai/connector/healthspan-intel
- Category: Health
- Developer: Longevity Dojo LLC (https://www.healthspanintel.com)
- Muse status: Waiting for Meta. Sent to Meta for approval. Until then you can add it with a link.
- Health: Working, 256ms, checked 2026-10-29T12:15:49Z
- Endpoint: https://www.healthspanintel.com/api/mcp
- Auth: No account needed; Pricing: free
- Screening: Screened, no issues found (2026-10-28T12:00:28Z)
- Source: Submitted by its builder and screened before listing

Healthspan Intel is a fact-checked longevity research service. The connector gives Muse read-only access to its published library: evidence cards with PubMed IDs and evidence tiers, protocol and head-to-head comparison verdicts, daily research briefings, and podcast episode summaries with verbatim quotes. Every response includes a canonical source URL and a medical disclaimer. Nothing is generated on demand or personalised. No sign-in or API key required.

How to connect: Not in Muse's Connectors list yet, but Muse can still use it. Paste this into Muse: "Use Healthspan Intel to help me. It is a free service with an MCP server at https://www.healthspanintel.com/api/mcp. It does not need an API key. Ask me before you share anything with it." Muse asks before it shares anything with the app's site. Meta does not review apps used this way, so only use ones you trust. We tested this in the Muse app on September 24, 2026: Muse used an app's link directly this way and returned a live answer.

Tools:
- search_evidence: Use when the user asks whether an intervention works, is safe, how much to take, or what research says. Accepts natural-language questions. Returns ranked stored papers with evidence tier, PMID and reviewed finding when available. It does not generate medical conclusions.
- get_evidence_card: Use when the user asks about a PMID or says tell me about this study. Returns title, abstract, authors, evidence tier, reviewed findings and limitations when available, retraction flags, newsletter mentions and canonical URL. Stored, reviewed content only; nothing generated or pe
- get_papers: Use when the user wants citation details for up to 25 PMIDs. Returns titles, authors, journals, dates, publication types, DOIs, citation counts and evidence library URL. No full text or new claims. Stored, reviewed content only; nothing generated or personalised.
- list_answer_capsules: Use when the user asks what answer capsules or reviewed topic answers are available. Returns capsule slugs, topic titles and canonical URLs. Stored, reviewed content only; nothing generated or personalised.
- get_answer_capsule: Use when the user asks for a concise reviewed answer on a covered longevity topic. Returns stored answer variants, confidence, evidence references, query tags, update date, version and canonical URL. Stored, reviewed content only; nothing generated or personalised.
- get_evidence_graph: Use when the user asks which studies support a topic's claims, how strong the evidence is, or what remains unknown. Returns claims, verdicts, confidence, supporting evidence, caveats, known unknowns and canonical URL. Stored, reviewed content only; nothing generated or personalis
- latest_newsletter: Use when the user asks what's new or wants today's brief. Returns the latest sent newsletter's title, plain text as sent, send date, word and source counts, and canonical URL; detail=full adds the stored draft and email HTML. Stored, reviewed content only; nothing generated or pe
- get_newsletter: Use when the user asks to read a specific newsletter or past brief by slug or ID. Returns its title, plain text as sent, send date, word and source counts, and canonical URL; detail=full adds the stored draft and email HTML. Stored, reviewed content only; nothing generated or per
- search_newsletters: Use when the user asks what newsletters covered about a longevity question. Accepts natural-language questions. Returns sent titles, subtitles, preheaders, archive metadata and URL. No drafts or internal fields. Stored, reviewed content only; nothing generated or personalised.
- list_protocols: Use when the user asks what protocols are available. Returns protocol slugs, titles, categories and canonical URLs. Stored, reviewed content only; nothing generated or personalised.
- get_protocol: Use when the user wants a protocol for VO2max, insomnia, muscle, time-restricted eating, Mediterranean diet or creatine. Returns verdict, evidence planks, tiers, caveats, PMIDs, review metadata and canonical URL. Stored, reviewed content only; nothing generated or personalised.
- list_comparisons: Use when the user asks what comparisons are available or which supplements have been compared. Returns comparison slugs, titles and canonical URLs. Stored, reviewed content only; nothing generated or personalised.
- get_comparison_verdict: Use when the user asks X vs Y, which is better, or how two supplements compare. Returns the comparison verdict, evidence tier, PMIDs, claim reviews, section IDs and canonical URL. Stored, reviewed content only; nothing generated or personalised.
- list_experts: Use when the user asks which experts are covered. Returns published expert names, slugs and canonical URLs. Unpublished pages are excluded. Stored, reviewed content only; nothing generated or personalised.
- get_expert_verdict: Use when the user asks what Attia, Huberman or Sinclair recommends or how an expert's protocol compares with evidence. Returns published expert verdict, protocol claims, evidence tiers, PMIDs and canonical URL. Stored, reviewed content only; nothing generated or personalised.
- list_recent_evidence: Use when the user asks for new longevity evidence or updates since a date. Returns evidence-card metadata and stored reviewed findings published since that date, newest first, with review timestamps and canonical URLs. Nothing generated or personalised by this tool.
- list_recent_updates: Use when the user asks for new longevity evidence or updates since a date. Returns sent newsletters, served protocol reviews and public episodes since that date, newest first, with canonical URLs. It excludes drafts, transcripts and claim groups. Nothing generated or personalised
- list_podcast_sources: Use when the user asks which podcasts are covered. Returns active podcast IDs, names, websites and categories, plus the podcast library URL. It does not return podcast transcripts or inactive sources. Stored, reviewed content only; nothing generated or personalised.
- search_episodes: Use when the user wants episodes on a topic. Returns titles, summaries, dates, URLs and pagination. Summaries and takeaways are AI-generated, not reviewed claims; quotes are verbatim. No transcripts or claim groups. Stored content only; nothing generated or personalised by this t
- get_episode: Use when the user asks about an episode. Returns summary, quotes, timestamps, takeaways, guests, PMIDs and audio link. Summaries and takeaways are AI-generated, not reviewed claims; quotes are verbatim. No transcripts or claim groups. Nothing generated or personalised by this too
- get_verdict: Use when the user asks whether an intervention works, is safe, or which of two is better. Returns the canonical stored verdict sentence, evidence tier, caveats, PMIDs and canonical URL. Stored, reviewed content only; nothing generated or personalised.
- search_verdicts: Use when searching reviewed verdicts for a named intervention or exact verdict title. Eligibility uses titles and declared topics; category-framed searches otherwise abstain. Returns stored verdicts with tiers, caveats, PMIDs and URLs. Reviewed content only; nothing generated or 
- search: Use when the user asks a longevity question and wants sources. Returns reviewed verdicts first, then experts, evidence cards, capsules, protocols, comparisons, newsletters and episodes, with IDs, titles and canonical URLs. Stored content only; nothing generated or personalised by
- fetch: Use when the user asks to read a result found by search. Returns its ID, title, stored fields as plain text, available evidence metadata and canonical URL. It does not return podcast transcripts. Stored content only; nothing generated or personalised by this tool.
- subscribe: Use when the user explicitly asks to subscribe to Healthspan Intel by email. Starts email verification and returns status and the reader's next step. No subscriber IDs or tokens. Stored, reviewed content only; nothing generated or personalised.
- search_scoped_evidence: Use when a reader asks about a published, scoped intervention or outcome. Returns question IDs and distinct source publication/review dates, or an explicit gap. Does not generate research, infer freshness or grade evidence by match rank.
- get_scoped_evidence: Use when a published scoped question ID needs its stored brief. Returns source-reported findings, versions, review dates and caveats. Does not generate new interpretation or personalized advice.
- inspect_scoped_support: Use when a published finding needs its retained support location and source rights. Restricted wording appears only as a hash and redaction reason. Does not generate evidence or retrieve full articles.
- compare_scoped_evidence: Use when two to eight findings from one published question need stored context and study dependence. Returns their source and rights context. Does not generate a treatment ranking or new synthesis.
- get_scoped_package: Use when a published question needs its exact versioned package, including support, gaps and revisions. Rights or reassessment changes withhold release. Does not generate research or a private investigation.

Screening checks:
- MCP handshake: pass (Answered in 165ms)
- Domain against threat feeds (Cloudflare security DNS): pass (www.healthspanintel.com not flagged)
- Published packages against the OSV malicious-package database: n/a (No npm or PyPI package published)
- Hidden instructions or invisible characters in tool text: pass (30 tools read, nothing found)
- Inputs asking for passwords, card numbers or seed phrases: pass (None found)
- Domain and redirects: pass (Domain registered 260 days ago)
- AI review of purpose and tool behavior: n/a (Automated checks only)