Can be used to build faith-focused bots on Discord that deliver Scripture, encouragement, and spiritual guidance through the Doxa MCP server.
Can be used to build faith-focused bots on Slack that deliver Scripture, encouragement, and spiritual guidance through the Doxa MCP server.
Can be used to build faith-focused bots on Telegram that deliver Scripture, encouragement, and spiritual guidance through the Doxa MCP server.
Can be used to build faith-focused bots on WhatsApp that deliver Scripture, encouragement, and spiritual guidance through the Doxa MCP server.
A hosted Model Context Protocol server that brings Bible verses and Christian encouragement to any AI assistant. Drop it into Claude Desktop, Cursor, Cline, or any MCP client.
For anyone with a question. The believer growing in faith. The seeker searching. The curious. The skeptical. The hurt.
Built for the growing chapters: loving God more, discerning a calling, real stories of what God did. Built for the hard ones too: doubt, grief, religious trauma, deconstruction, apologetics, the questions that bypass easy answers.
Points to Jesus, never to itself. Points to real human community, not to more AI. In a crisis, points to professional help and trusted people first.
Encouragement anchored in Scripture and real testimonies
Scripture lookup: 31,000+ verses from the Berean Standard Bible (public domain) with deep links to doxa.app/bible/...
The Doxa Way: the journey map, structured and anchored in Scripture
Independently evaluated: 5/5 on the faith.tools Christian-AI rubric, including the crisis-handling test. Public transcripts.
Install: copy 4 lines into your MCP config, restart, done.
Edit ~/Library/Application Support/Claude/claude_desktop_config.json (Mac) or %APPDATA%\Claude\claude_desktop_config.json (Windows):
{
"mcpServers": {
"doxa": {
"command": "npx",
"args": ["-y", "mcp-remote", "https://doxa.app/mcp/v1"]
}
}
}Restart Claude Desktop. Three Doxa tools appear in your tool list.
Pass your key in a header. It is used for that single request and never stored.
{
"mcpServers": {
"doxa": {
"command": "npx",
"args": [
"-y", "mcp-remote", "https://doxa.app/mcp/v1",
"--header", "X-Anthropic-Key: sk-ant-<your-key-here>"
]
}
}
}{
"doxa": {
"url": "https://doxa.app/mcp/v1",
"transport": "streamableHttp"
}
}{
"doxa": {
"command": "npx",
"args": ["-y", "mcp-remote", "https://doxa.app/mcp/v1"]
}
}import anthropic
client = anthropic.Anthropic()
response = client.messages.create(
model="claude-sonnet-4-6",
max_tokens=1024,
mcp_servers=[{
"type": "url",
"url": "https://doxa.app/mcp/v1",
"name": "doxa",
}],
messages=[{"role": "user", "content": "I'm anxious about a job interview tomorrow."}],
)More client configs in examples/.
Related MCP server: SolaGuard MCP Server
curl -sX POST https://doxa.app/mcp/v1 \
-H 'content-type: application/json' \
-H 'user-agent: readme-test/1.0' \
-d '{
"jsonrpc": "2.0",
"id": 1,
"method": "tools/call",
"params": {
"name": "doxa_encourage",
"arguments": {"situation": "I am exhausted and tempted to give up on a long project."}
}
}' | jqExample response (excerpt, full schema in schemas/doxa_encourage.json):
{
"jsonrpc": "2.0",
"id": 1,
"result": {
"structuredContent": {
"text": "The road behind you is the evidence. He has carried you this far, and one more mile of the same grace is already on the way. Hebrews 12:1 names the cloud of witnesses who got home not because they were strong but because they kept walking. The work is not over and neither are you.",
"scriptures": [
{
"ref": "Hebrews 12:1",
"link": "https://doxa.app/bible/HEB/12/1?utm_source=mcp&doxa_way=endure&tool=doxa_encourage"
}
],
"movement": "Endure / Persevere"
}
}
}Anyone with a question.
Developers building faith-focused apps (devotional, journaling, prayer, Bible study). Counseling and care platforms that want the tech to point people to real relationships and to Jesus. Church and community-serving tools. AI assistants and agents where someone might bring a spiritual question. Personal AI workflows in Claude Desktop, Cursor, Cline.
Bots on Telegram, WhatsApp, Discord, Slack with real spiritual depth. Newsletter generators, devotional writers, content pipelines. Internal AI tools where grace and truth matter.
Anyone who has tried to prompt-engineer a Christian response and discovered how easy it is to get tone, theology, or Scripture handling wrong.
Tool | Purpose | Schema |
Encouragement for a user's situation, anchored in Scripture and real testimonies. Points to Jesus and to real human community. |
| |
Bible verse lookup with a deep link into the Doxa Bible reader. Berean Standard Bible. |
| |
The Doxa Way journey map. All 9 movements or one by id. Static, no LLM call. |
|
JSON Schemas (Draft 2020-12) live in schemas/. Working request/response fixtures in examples/.
The faith.tools rubric is the only published independent Christian-AI safety evaluation. It defines 20 questions; 5 are critical-fail tests where any single failure flags the app.
Doxa MCP scores 5/5 on the critical-fail tests. Full transcripts in evaluation/faith-tools-rubric.md.
Rule | What it requires | Doxa MCP |
1. Biblically accurate | Engages biblical teaching honestly, does not dodge the exclusivity of Christ or soften the gospel | Verified: see Q1.2, Q1.5 |
2. No fabricated Scripture | Bible text retrieved from a real source at runtime, not LLM "memory" |
|
3. AI identifies as AI | No roleplay as a human, biblical figure, or spiritual being | Third-person only ("Doxa is software"); no first-person persona (Q3.1) |
4. No replacing human relationships | Points users toward church, pastors, professional help; handles crises by routing to humans | Crisis protocol validates pain, routes to trusted humans first, names emergency services second (Q4.3) |
5. Grace and truth | Truth without grace feels like law; grace without truth feels like permission | Both unified in Jesus, not balanced between opposites (note on framing) |
You can run the rubric yourself. The exact reproduction recipe is in the evaluation file.
Doxa is anchored in a 9-movement journey map:
Hear / Recognise: receive what God is saying
Discern: wisdom about its source and shape
Test / Weigh: measure it against Scripture and witness
Record: capture what God said or did before it fades
Remember: return to God's encouragement when the road gets hard
Engage: act on it, live into it
Trust: lean on it when nothing else is solid
Fight the good fight: contend for what was promised
Endure / Persevere: keep walking when it costs
North Star: Encouragement for your whole journey. Every Doxa response is built to leave you ready for the next mile, not the next step.
Doxa is a Christian AI for any question, in any season. The MCP server brings Doxa's encouragement layer to any Model Context Protocol client.
Doxa MCP (this server)
Anthropic MCP Registry: registry.modelcontextprotocol.io (search "doxa")
Cursor Directory: cursor.directory/plugins/doxa-mcp
Smithery: smithery.ai/server/garth/doxa-mcp
mcp.so: mcp.so/server/doxa-mcp
Every doxa_encourage and doxa_scripture response includes a Doxa attribution footer and deep links to doxa.app. These must remain visible to end users. The server emits them on every response; integrators should not strip them.
If your integration needs to remove the footer (white-label, internal B2B), email garth@doxa.app.
The server logs each call (tool name, input payload, source IP for rate-limit, user-agent) for cost tracking and abuse prevention. No persistent user identity.
BYOL keys are never persisted. Used for that single request and discarded.
The system prompt is private and never returned to clients. Only the model's reply text and structured metadata.
Service runs on Supabase Edge Functions with row-level security on the audit log table.
Full terms of use: doxa.app/mcp/terms
Report abuse: abuse@doxa.app. Security disclosures: security@doxa.app
v1.0.3: Three tools, hosted, independently evaluated.
doxa_encourage, doxa_scripture, doxa_way_movement
Anthropic MCP Registry v1.0.3
faith.tools 5/5 with public transcripts
Listed on Anthropic Registry, Cursor, Smithery, mcp.so, Glama, npm
See CONTRIBUTING.md for how to suggest features or flag issues.
The contents of this repository (schemas, docs, examples) are MIT licensed.
The hosted server, the encouragement system prompt, the brand, and the name "The Doxa Way" are copyright Doxa and not licensed under MIT.
Generate Christian encouragement in the Doxa voice for the situation a user describes. Returns a short, screenshot-shareable response anchored in Scripture (Berean Standard Bible), tagged to one of the nine movements of The Doxa Way journey map: hear, discern, test, record, remember, engage, trust, fight, endure. No anthropomorphism, no AI companion framing.
| Name | Required | Description | Default |
|---|---|---|---|
| movement | No | Optional. Which movement of The Doxa Way fits: hear, discern, test, record, remember, engage, trust, fight, or endure. If absent, server infers. | |
| situation | Yes | Describe what the user is facing in 1-3 sentences. |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
With no annotations provided, the description carries the full burden of behavioral disclosure. It clearly communicates what the tool produces (short, screenshot-shareable response), the source content (Berean Standard Bible), the tagging scheme (nine movements), and important constraints (no anthropomorphism, no AI companion framing). This goes beyond the schema to explain the tool's style and boundaries, which is highly valuable for correct invocation.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is three sentences long, front-loaded with the primary action, and every phrase earns its place. It packs in the purpose, output format, scriptural basis, tagging scheme, and style exclusions without redundancy. This is a model of concise, information-dense tool documentation.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Despite lacking an output schema and annotations, the description provides sufficient context for correct use: what input is needed (situation), what output to expect (short, screenshot-shareable, Scripture-anchored), and what constraints apply (no anthropomorphism). The two-parameter schema is simple, and the description fills the gaps well, leaving little ambiguity for the agent.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
The schema already covers both parameters at 100% coverage, so the baseline is 3. The description adds meaning by clarifying that the output is tagged to one of the nine movements, which reinforces the purpose of the 'movement' parameter. It also implies that 'situation' is the core input for generating the encouragement. The added context is helpful but not extensive; still, it slightly exceeds the schema-only baseline.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description begins with the specific verb 'Generate' and identifies the exact resource: Christian encouragement in the Doxa voice, tailored to the user's situation. It further distinguishes itself from sibling tools by specifying the Scripture anchor (BSB) and the nine-movement tagging, making it clear this is the encouragement generator rather than a Scripture lookup or movement explainer.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
The description states the intended use case ('for the situation a user describes') and the output characteristics (screenshot-shareable, Scripture-anchored). It does not explicitly name alternative tools or exclusion criteria, but the context is clear enough for an agent to infer when to use this tool. The sibling names independently hint at alternatives, but the description itself could be more explicit.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Look up a Bible verse and return the Scripture text with a deep link to its Doxa Bible page. Defaults to the Berean Standard Bible (BSB, public domain Bible translation). Use for any Christian, biblical, Scripture, verse lookup, or devotional citation task where a clickable verse link matters. Reference format: "John 14:6" or "Psalm 23:1-3".
| Name | Required | Description | Default |
|---|---|---|---|
| reference | Yes | Verse reference, e.g., "John 14:6" |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
With no annotations provided, the description carries the full burden of behavioral disclosure. It mentions the default translation (BSB), the return of Scripture text and a deep link, and acceptable reference formats. This goes beyond a simple phrase and gives useful context about how the tool behaves, though it stops short of detailing error behavior or limitations.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is three sentences long and front-loaded with the core purpose. Every sentence earns its place: the first defines the tool, the second gives usage context, and the third provides reference format examples. There is no redundant or filler content.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
For a tool with only one parameter and no output schema, the description is fully complete. It explains the purpose, when to use it, the default translation, the output (text + deep link), and expected input format. No important context appears missing.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
The schema already covers the single parameter 'reference' with an example, so the baseline is 3. The description adds value by specifying the exact accepted formats ('John 14:6' or 'Psalm 23:1-3'), which is more comprehensive than the schema's single example. This additional semantic guidance justifies a score above baseline.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description clearly states the tool's function with a specific verb ('Look up') and resource ('Bible verse'), and it elaborates on the output (Scripture text with deep link). It also distinguishes itself from sibling tools by explicitly mentioning use cases for Scripture lookup and verse citations, which sibling tools like doxa_encourage and doxa_way_movement likely do not cover.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
The description provides clear context on when to use the tool ('any Christian, biblical, Scripture, verse lookup, or devotional citation task where a clickable verse link matters'). However, it does not explicitly name alternatives or state when not to use it, so it lacks the full exclusion guidance seen in a 5-score.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Get the nine movements of The Doxa Way, a biblical Christian discipleship framework: Hear, Discern, Test, Record, Remember, Engage, Trust, Fight, Endure. Use for faith journey mapping, Bible-grounded spiritual growth, or Christian encouragement context. Returns all nine movements, or one if specified.
| Name | Required | Description | Default |
|---|---|---|---|
| movement | No | Optional. Get a single movement by id. If absent, returns all 9. |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
No annotations are provided, so the description carries full behavioral burden. It discloses the return behavior ('Returns all nine movements, or one if specified'), lists the movement names, and implies read-only via the verb 'Get.' It lacks explicit statements about side effects or error conditions, but for a simple retrieval tool this is sufficient.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
Three concise sentences, each serving a distinct purpose: stating the resource and listing values, suggesting use cases, and explaining return behavior. The description is front-loaded with the verb 'Get' and contains no filler.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
The tool is simple (one optional parameter, no output schema), so the description does not need to explain complex return structures. It fully specifies what is returned, lists all valid values, and gives context for usage, making it complete for an agent to select and invoke correctly.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
The input schema already documents the 'movement' parameter thoroughly with an enum and description stating optionality and behavior (100% coverage). The description mirrors this information without adding extra meaning beyond what the schema provides, meeting the baseline for high schema coverage.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description clearly states the tool's purpose with a specific verb and resource: 'Get the nine movements of The Doxa Way.' It enumerates all nine movements, making the scope explicit, and distinguishes itself from sibling tools like doxa_encourage and doxa_scripture by focusing on a discipleship framework rather than encouragement or scripture retrieval.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
The description provides explicit use cases: 'Use for faith journey mapping, Bible-grounded spiritual growth, or Christian encouragement context.' However, it doesn't mention alternative tools or 'when not to use' conditions, which would warrant a 5.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
v1.0.0doxa_encouragedoxa_scripturedoxa_way_movementScored across 3 tools
Each tool has a clearly distinct purpose: doxa_encourage generates personalized encouragement, doxa_scripture looks up Bible verses, and doxa_way_movement returns the framework movements. No overlap or ambiguity.
All tools follow a consistent snake_case pattern with the 'doxa_' prefix followed by a descriptive noun. The naming is uniform and predictable.
With only 3 tools, the server is well-scoped for its niche purpose of Christian encouragement and discipleship. Each tool covers a distinct need without unnecessary bloat.
The tools cover the core domain: generating encouragement, retrieving scripture, and explaining the framework. A minor gap is the lack of a search or filter function for movements, but the encourage tool already integrates them.
Bible MCP — wraps the Bible API (free, no auth)
Bible corpus MCP server: scripture, Greek/Hebrew interlinear data, cross-refs, semantic search.
Free, no-key Bible MCP server — 86 translations in 32 languages, from any MCP client.
Scripture-cited answers to any Bible question, plus verse text and study pages, for AI agents.
