bryan-protocol-mcp
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| PYTHONPATH | No | Path to the repository's `src` directory, used when running the server from source (e.g. `/path/to/bryan-protocol-mcp/src`). Not needed if the package is installed in editable mode. |
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
You are querying the Bryan Johnson / Blueprint Ground Truth Archive. Always prioritize Tier 1 (protocol.bryanjohnson.com) and explicit stopped or avoid stances over older videos or external interviews. Always cite exact URLs, dates, and video timestamps.
Capabilities
Features and capabilities supported by this server
Protocol revision2025-11-25
| Capability | Details |
|---|---|
| tools | {
"listChanged": false
} |
| prompts | {
"listChanged": false
} |
| resources | {
"subscribe": false,
"listChanged": false
} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| get_protocol_overviewA | Get database stats, source tiers, last-ingested metadata, and the index of all official Tier 1 protocol.bryanjohnson.com sections. |
| get_official_protocolB | Look up authoritative Tier 1 sections from protocol.bryanjohnson.com for a given topic (e.g. sleep, nutrition, exercise, skin, oral care, hair, biomarkers, rx). |
| what_would_bryan_doA | Retrieve a tiered Ground Truth Dossier (explicit stances, Tier 1 protocol pages, Tier 2 Bryan videos/articles, Tier 3 external podcast timestamped quotes) to answer what Bryan Johnson would or wouldn't do in a scenario. |
| check_stance_and_changesA | Check whether Bryan actively uses/does something, stopped doing it, or strictly avoids it, including rationale and chronological mentions across videos and articles. |
| search_archiveC | Full-text BM25 search across all official protocol pages, Blueprint blog articles, @BryanJohnson videos/streams, and external podcast transcripts. |
| get_documentA | Fetch the full text (and optionally timestamped segments) of any document ID returned by search or list_sources. |
| list_sourcesC | List ingested documents, articles, videos, and podcasts with pagination and optional tier/source_type filters. |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
| evaluate_against_blueprint | Evaluate a food, supplement, habit, or daily routine strictly against Bryan Johnson's documented Blueprint Protocol. |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
| Official Protocol Index | Index of all Tier 1 protocol.bryanjohnson.com pages and archive statistics. |
| Bryan Stances & Anti-Protocol Matrix | Structured ground-truth stances (active, stopped, avoid, experimental) with citations. |
TDQS
Scored across 7 tools
Several tools target overlapping questions about Bryan's positions: what_would_bryan_do, check_stance_and_changes, get_official_protocol, and search_archive can all plausibly answer 'what does Bryan think about X'. The descriptions differentiate them somewhat (scenario dossier vs. chronological stance tracking vs. Tier 1 only vs. broad search), but an agent could still misselect among the four.
Most names follow a clean verb_noun snake_case pattern (get_protocol_overview, get_official_protocol, check_stance_and_changes, search_archive, get_document, list_sources). The outlier is the question-style 'what_would_bryan_do', which breaks convention but is memorable and domain-appropriate.
Seven tools is well within the ideal 3-15 range for a knowledge-retrieval server. Each tool has a plausible distinct role (overview, targeted lookup, scenario dossier, stance tracking, search, fetch, list), so nothing feels padded.
The surface covers the core lifecycle of a RAG-style knowledge server: discover (overview/list_sources), search (search_archive), and fetch (get_document), plus specialized protocol/stance lookups. Minor gaps exist (e.g., no explicit topic/section enumeration tool beyond the overview, no related-documents or cross-source comparison), but agents can work around these via search_archive and get_document.