Skip to main content
Glama
datacore-one

datacore-mcp

Official
by datacore-one
OverviewSchemaSourceRelated ServersDiscussionsAdmin
TypeScript
Hybrid

@datacore-one/mcp

CI

The knowledge and task orchestration layer for AI assistants — journals, GTD, Zettelkasten, and autonomous workflows over plain-text files via MCP.

Why

AI assistants are great at reasoning but have nowhere to put what matters: your decisions, your tasks, your notes.

Datacore gives them a structured, plain-text home — capture journal entries, manage GTD task lists, build a Zettelkasten, and extend with modules for autonomous overnight workflows.

Persistent memory — engrams, learning, and recall — is handled by Datacore's companion server, PLUR (plur_* tools). Run the two side by side: PLUR remembers, Datacore organizes.

Not a RAG system. Not a vector database you have to manage. Just plain-text files and an MCP server.

Related MCP server: Facthouse

Quick Start

Install globally:

npm install -g @datacore-one/mcp

Then connect from any MCP-compatible client. On first use, the server creates ~/Datacore/ with:

  • journal/ — Daily session logs

  • knowledge/ — Ingested reference material

  • engrams.yaml — Shared engram store, read and written by the companion PLUR MCP

  • packs/ — Engram packs used by PLUR

  • config.yaml — Configuration (all fields optional)

  • CLAUDE.md, AGENTS.md, .cursorrules, .github/copilot-instructions.md — Editor context files so any AI assistant immediately understands Datacore

Everything is plain text -- no databases, no lock-in.

Connecting

Datacore is a standard MCP server. It works with any client that speaks MCP v1.0+ over stdio or HTTP -- the AI model behind the client does not matter.

Claude Code

Add to .mcp.json in your project root (or ~/.claude.json globally):

{
  "mcpServers": {
    "datacore": {
      "command": "datacore-mcp"
    }
  }
}

Then allow Datacore tools in .claude/settings.json (or .claude/settings.local.json):

{
  "permissions": {
    "allow": [
      "mcp__datacore"
    ]
  },
  "enableAllProjectMcpServers": true
}

This auto-approves all Datacore MCP tools (capture, search, status, etc.) so you don't get prompted on every call. The enableAllProjectMcpServers setting ensures the MCP server defined in .mcp.json is activated automatically.

Claude Desktop

Add to ~/Library/Application Support/Claude/claude_desktop_config.json:

{
  "mcpServers": {
    "datacore": {
      "command": "datacore-mcp"
    }
  }
}

Cursor / Windsurf / Other MCP Clients

Most MCP-compatible editors use the same config format. Check your editor's MCP documentation for where to place the server config. The command is always datacore-mcp.

HTTP (Remote / Multi-Client)

For shared or remote setups, run in HTTP mode:

datacore-mcp --http

Then point your MCP client to http://127.0.0.1:3100/mcp. See HTTP Transport for options.

Two Modes

Mode

Storage

What You Get

Core (~/Datacore)

Flat files

Journal, knowledge, dates, packs

Full (~/Data)

Datacore system

+ modules, GTD, spaces, Datacortex

Mode is auto-detected. If you have a full Datacore installation at ~/Data, it uses that. Otherwise it creates a lightweight ~/Datacore directory.

Override with environment variables: DATACORE_PATH (full) or DATACORE_CORE_PATH (core).

Tools (5 core + 3 full-mode)

Datacore exposes productivity tools. Memory — engrams, learning, recall, packs — is provided by the companion PLUR MCP server (plur_* tools), not by Datacore.

Core

Tool

Description

datacore_capture

Write a journal entry or knowledge note

datacore_search

Search journal and knowledge by keyword or semantic

datacore_ingest

Ingest text as a knowledge note

datacore_status

System status, counts, actionable recommendations

datacore_date

Canonical date operations (today, day-of-week, validate, add/sub, parse, org-stamp)

Modules (full mode only)

Tool

Description

datacore_modules_list

List installed modules

datacore_modules_info

Detailed info about a module

datacore_modules_health

Health check for modules

Tool names use underscores to satisfy the MCP tool-name rule ^[a-zA-Z0-9_-]{1,64}$. Legacy dot-namespaced names (datacore.capture) are still accepted as aliases for backward compatibility.

Managed installations select exactly one absolute DATACORE_PATH (existing full installation) or DATACORE_CORE_PATH (existing or new core store). Invalid explicit paths fail without choosing a different store. Leave both unset only when the documented HOME-based discovery is intended.

Set absolute DATACORE_LIB and DATACORE_PYTHON to the qualified core library and interpreter. Failed explicit selections do not fall back to mutable data code or another Python. Ledger status uses the installed ledger_health.py version 1 protocol and canonical space discovery. Missing helpers, unreadable spaces, busy writers or invalid responses remain unverified and cannot yield “System healthy.” Older cores need reconciliation before ledger health can be established. Datacortex uses the same interpreter and the module next to that installed library, with bounded foreground process cleanup; deployment remains responsible for independent OS and credential isolation.

Full-mode space discovery requires the installed space_catalog.py version 1 helper. It calls the core spaces.py implementation (DIP-0015) and includes root, named, nested and canonical legacy spaces. Missing or malformed discovery refuses startup. Personal capture/ingestion and global module data require one unambiguous personal space; they never default to a team space. Scoped module names use the stable marker name and data paths use its actual directory. After a space identity or routing path changes, restart the server; stale sessions refuse tool calls. Existing data is not moved or renamed by discovery. The earlier audit's unpublished ordinal-based scoped tool names are replaced by datacore_<stable-space-name>_<module>_<tool>; global names remain unchanged.

Journal resources validate calendar dates and read bounded, unlinked regular files. Keyword search reads current source files without retaining a process content cache; an index in one space cannot hide matches in another. Linked, changing, oversized or unreadable files and scan limits yield an explicit incomplete-coverage warning. Files are limited to 4 MiB, with a 32 MiB content budget, 10,000 directory entries and depth 32 per keyword search.

Capture and ingestion create private notes with unique filenames and complete, non-replacing publication; existing note filenames and contents are preserved. Journal capture appends without rewriting earlier entries and syncs before acknowledging success. Local MCP writers and initializers coordinate through SQLite in state/mcp-file-writes/coordination.db; process death releases that lock. Keep this machine-local state out of synchronization. Initialization publishes complete defaults and packs without replacing existing user files. Aliased write directories and linked mutable journals are refused.

These write checks are qualified on macOS/Linux filesystems. They do not provide cross-host locking or an exactly-once retry protocol. An interrupted request may have left a complete note or a partial new journal entry; inspect the destination before retrying a request whose durability could not be confirmed. Incomplete .datacore-pending-* artifacts are private and are not acknowledged notes.

Prompts

The server provides MCP prompts — workflow templates your AI can discover and use automatically:

Prompt

Description

datacore-capture-guide

Capture a journal entry or knowledge note

datacore-guide

Complete guide to Datacore tools and workflows

Prompts are the primary way the AI understands Datacore. When your AI connects, it can list available prompts and immediately knows how to capture, search, and organize — and that persistent memory lives in PLUR.

Resources

Resource

Description

datacore://guide

Agent workflow reference (markdown)

datacore://status

System status summary (JSON)

datacore://journal/today

Today's journal entry (markdown)

datacore://journal/{date}

Journal entry by date

Memory (via PLUR)

Datacore organizes; PLUR remembers.

Persistent memory — engrams, learning, recall, feedback, and engram packs — lives in the companion PLUR MCP server (plur_* tools). Datacore scaffolds the shared, plain-text data directory (including engrams.yaml and packs/) that PLUR reads and writes, so both servers work against the same ~/Data or ~/Datacore store.

Connect both in your MCP client and your AI gets structured knowledge and task management (Datacore) plus persistent memory (PLUR). See the PLUR docs for the memory toolset and engram lifecycle.

Upgrading from ≤1.5? The engram engine (learn, inject, recall, promote, feedback, forget, packs, and the engagement/XP layer) moved out of Datacore into PLUR. Install @plur-ai/mcp alongside Datacore to keep that functionality.

Configuration

Environment Variables

Variable

Default

Description

DATACORE_PATH

~/Data

Full installation path

DATACORE_CORE_PATH

~/Datacore

Core mode storage path

DATACORE_TIMEZONE

System

IANA timezone (e.g., Europe/Ljubljana)

DATACORE_LOG_LEVEL

warning

debug, info, warning, error

DATACORE_CACHE_TTL

60

File cache TTL in seconds

DATACORE_TRANSPORT

stdio

stdio or http

DATACORE_HTTP_PORT

3100

HTTP transport port

DATACORE_HTTP_HOST

127.0.0.1

HTTP bind address

config.yaml

Create config.yaml in your Datacore directory (or .datacore/config.yaml in full mode):

version: 2
search:
  max_results: 20
  snippet_length: 500        # chars around match
hints:
  enabled: true              # include _hints in tool responses for agent guidance

All fields have defaults -- the file is optional. Memory-related settings (engrams, packs, engagement) are configured in PLUR, not here.

HTTP Transport

For remote or multi-client setups:

DATACORE_HTTP_PORT=8080 datacore-mcp --http

  • MCP endpoint: POST /mcp

  • Health check: GET /health

  • Default bind: 127.0.0.1:3100

Module System (Full Mode)

Full Datacore installations discover module tools from .datacore/modules/ and [space]/.datacore/modules/. Modules ship executable tools/index.js; discovery does not compile TypeScript.

Installation

Callable name

Default data directory

Global crm

datacore_crm_lookup

0-personal/.datacore/module-data/crm/data/

Personal crm

datacore_0-personal_crm_lookup

0-personal/.datacore/module-data/crm/data/

Team crm

datacore_1-team_crm_lookup

1-team/.datacore/module-data/crm/data/

Scope is part of each space module's callable identity. Calls never choose a data destination by discovery order or fall back to a module in another space. Third-party module acme/crm uses namespace acme-crm; its private data directory uses the manifest name (.../module-data/acme/crm/data/). Module code directories may use the flattened name acme-crm.

Upgrade: update callers of space-installed tools to the name advertised by tools/list. Their old unqualified names have no implicit alias, since such an alias could silently select a different space. Existing global callable names remain stable. Duplicate names (including collisions with core tools) and invalid or overlong identifiers are refused; unrelated tools remain available. Names must fit the 64-character MCP limit.

Private data now has a separate root from installed module code. Legacy data, state, or settings.local.yaml in either the installed code or historical scoped directory blocks that module with module-data-unverified. Stop its writers, preserve backups, and use core's module_data_migrate.py with the verified space identity and legacy source directory. The helper retains the originals in private backup and supports interrupted retries; an incomplete receipt keeps the module unavailable. See core's .datacore/lib/RUNTIME.md for the procedure and filesystem limits. Loading modules never moves old data or silently replaces it with an empty store. Verify real reads and ownership under the installed service identity before resuming it.

The full-mode server exposes all installed scopes to its owner. dataPath is a routing convention: trusted module handlers execute in the same process and retain its filesystem privileges. Use independently restricted processes, credentials and storage roots where separate security contexts are required.

Registration validates handlers and argument contracts. Zod 3, Zod 4 and supported JSON Schema tools retain input validation; raw JSON Schema cannot fetch remote references or silently coerce input. Health reports the actual startup registration snapshot for each installed scope, without importing modules again or copying raw exceptions. A name shared by multiple scopes cannot select one health result implicitly.

The @datacore-one/mcp/runtime export provides z, yaml and findPython from the selected package environment in ESM and CommonJS forms. findPython() uses the same Python 3.10+ selection as core discovery; an invalid explicit DATACORE_PYTHON returns no interpreter and never enables fallback. Delegates must still select installed code, bound their requests and subprocess lifetime, and restrict the child environment. The module must first be able to resolve the MCP package through an explicit installed package binding. A global installation or NODE_PATH alone does not make an ESM import resolve. Qualify that binding from the module's physical directory and service identity, or ship a qualified module bundle. DIP-0049 is a draft design discussion, not a claim that its entire installation proposal has been implemented.

License

MIT

Development verification

The full release gate exercises both standalone mode and the actual core discovery provider. Check out the core commit pinned in .github/workflows/ci.yml, create a Python 3.10+ virtual environment, and install scripts/requirements-core-tests.txt with pip install --require-hashes --no-deps. Set DATACORE_LIB to that checkout's absolute .datacore/lib path and DATACORE_PYTHON to the virtual environment's absolute interpreter path. Run npm ci, npm run verify, and ./node_modules/.bin/tsc --noEmit. CI performs these steps in isolated directories and never uses an operator installation.

Available Tools

5 tools
datacore_captureA

Capture a journal entry or knowledge note. Call proactively to record important decisions, meeting outcomes, and significant events.

ParametersJSON Schema
NameRequiredDescriptionDefault
tagsNoTags to attach
typeYes
titleNoTitle for knowledge notes
contentYesContent to capture

TDQS

A3.5/5.0
Behavior2/5

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 implies a write operation ('capture', 'record') but does not explain side effects, persistence, permissions, or return values. The agent is left guessing what happens after the capture and whether it has any irreversible effects.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness5/5

Is the description appropriately sized, front-loaded, and free of redundancy?

The description is two concise sentences, front-loaded with the core purpose and followed by usage guidance. Every word adds value; there is no fluff or repetition of schema details.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness3/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

For a tool with 4 parameters and no output schema, the description covers the 'what' and 'when' but not the 'how' behaviorally. It doesn't mention what happens after capture (e.g., storage, retrieval via search), error conditions, or idempotency. It's adequate but leaves gaps in understanding the full workflow.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters3/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema description coverage is 75%, so the baseline is 3. The description adds little beyond the schema, only reinforcing the type distinction via 'journal entry or knowledge note'. It doesn't explain nuanced parameter usage (e.g., when title is required, how tags are used) beyond what the schema already provides.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose4/5

Does the description clearly state what the tool does and how it differs from similar tools?

The description clearly states the tool captures a journal entry or knowledge note, giving a specific verb and resource. It also provides examples of when to use it (decisions, meetings, events), which helps differentiate from sibling tools like search and status, though it doesn't explicitly contrast with datacore_ingest.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines4/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

The description gives clear proactive usage context: 'Call proactively to record important decisions, meeting outcomes, and significant events.' It does not explicitly state when not to use it or mention alternatives, but the guidance is sufficient for typical use cases.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

datacore_dateA

Canonical date operations — ALWAYS use this instead of typing dates from memory. LLMs hallucinate day-of-week names and anchor to training-era dates. Returns today's date, validates day-of-week, adds/subtracts days, parses relative expressions, and formats org-mode timestamps.

ParametersJSON Schema
NameRequiredDescriptionDefault
nNoNumber of days for add/sub
opYesOperation: today (current date+dow), dow (day-of-week for a date), validate (check date matches claimed dow), add/sub (N days from date), diff (days between two dates), parse (relative expression like "next monday"), org-stamp (<YYYY-MM-DD Day>)
dayNoClaimed day name (Mon..Sun) for validate
dateNoISO date YYYY-MM-DD (for dow/validate/add/sub/org-stamp)
exprNoRelative expression for parse (e.g. "tomorrow", "next monday", "in 3 days")
date2NoSecond date for diff
inactiveNoFor org-stamp: use [..] instead of <..>

TDQS

A4.4/5.0
Behavior4/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

No annotations are provided, so the description carries the full burden. It transparently states the tool's behaviors (returns today's date, validates day-of-week, adds/subtracts days, etc.) and explains the underlying problem it solves (LLM hallucination). While it doesn't explicitly mention side effects or read-only nature, the tool is inherently a pure calculation function, making the description sufficient.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness5/5

Is the description appropriately sized, front-loaded, and free of redundancy?

The description is two sentences, front-loaded with an imperative rule and rationale, followed by a compact enumeration of capabilities. Every word earns its place, with no redundancy or unnecessary details.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness4/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

For a simple date utility with one required parameter and fully documented schema, the description is largely complete. It does not specify return value formats (e.g., exact output of diff or parse), but the schema's op enumeration provides sufficient context. Since there is no output schema, slightly more detail on outputs would push this to a 5, but the low complexity keeps it adequate.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters3/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema description coverage is 100%, so the baseline is 3. The description enumerates operation categories but does not add parameter-level semantics beyond the schema. It does provide context on why certain operations exist (e.g., validation to counter hallucination), but this is not parameter-specific detail.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

The description clearly identifies the tool as a canonical date operations utility and enumerates all supported operations (today, dow, validate, add/sub, diff, parse, org-stamp). It distinguishes itself from sibling tools (datacore_capture, datacore_search, datacore_ingest, datacore_status) by being the only date-related utility.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines5/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

Explicitly instructs 'ALWAYS use this instead of typing dates from memory', providing a clear when-to-use directive with the rationale that LLMs hallucinate day-of-week names and training-era dates. This effectively tells the agent to prefer this tool for any date calculation, overriding memory-based guesses.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

datacore_ingestC

Ingest text content as a knowledge note

ParametersJSON Schema
NameRequiredDescriptionDefault
tagsNo
titleNo
contentYesContent to ingest

TDQS

C2.8/5.0
Behavior2/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

With no annotations, the description carries the full burden. It only states the action ('ingest') but does not disclose side effects, permissions, or what happens to the content once ingested. For a write operation, this lacks necessary behavioral detail.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness5/5

Is the description appropriately sized, front-loaded, and free of redundancy?

The description is a single succinct sentence with no filler words. It is appropriately sized for the tool's apparent simplicity and communicates the core purpose efficiently.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness2/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

The tool has no annotations, an output schema, and minimal schema coverage. The description omits information about how the content is stored, the role of tags and title, and any return values or side effects. Given the lack of annotations and output schema, this is incomplete for an agent to fully understand the tool's behavior.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters2/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema coverage is only 33%, with only 'content' described. The description mentions 'text content' but does not explain the 'tags' or 'title' parameters, nor does it compensate for the missing schema descriptions. It adds little beyond the parameter names.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose4/5

Does the description clearly state what the tool does and how it differs from similar tools?

The description clearly states the tool ingests text content as a knowledge note, using a specific verb and resource. However, it does not explicitly distinguish it from sibling tools like datacore_capture, which may perform a similar function.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines2/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

No guidance is provided on when to use this tool versus alternatives like datacore_capture. The description simply states the action without any contextual or exclusionary information.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

datacore_statusA

Show Datacore status: note counts, module health, update info

ParametersJSON Schema
NameRequiredDescriptionDefault

No parameters

TDQS

A4/5.0
Behavior3/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

With no annotations provided, the description carries the full burden of disclosing behavior. It states the information returned (note counts, module health, update info), which is helpful, but it does not explicitly confirm that the operation is read-only or mention any potential side effects, latency, or authentication requirements.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness5/5

Is the description appropriately sized, front-loaded, and free of redundancy?

The description is a single, front-loaded sentence that conveys the tool's purpose without any redundant information. Every word contributes to understanding what the tool does.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness4/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

For a simple, zero-parameter status tool with no output schema, the description adequately covers the key aspects: it states the resource (Datacore) and the specific status categories (note counts, module health, update info). While it could mention whether the status is real-time or cached, this is a minor gap and the description is sufficiently complete for its complexity.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters4/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

The tool has zero parameters, and the input schema is empty, so schema description coverage is trivially 100%. The description does not need to elaborate on parameter semantics, and the baseline for zero-parameter tools is 4, which is appropriate here.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

The description uses the specific verb 'Show' and resource 'Datacore status', explicitly listing what users can expect: note counts, module health, and update info. This clearly distinguishes the tool from sibling tools like datacore_capture, datacore_search, datacore_ingest, and datacore_date, which perform different operations.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines3/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

The description implies this tool is used for checking overall Datacore system health, but it does not explicitly state when to use it versus alternatives or when not to use it. With no competing status tools among siblings, the context is clear but the guidance is not explicit.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections.

  1. 5 tool updatesv2.1.0
    • Changeddatacore_capture1 field changed
      • changedInput schema / $schema
        Previous value: -"http://json-schema.org/draft-07/schema#"New value: +"https://json-schema.org/draft/2020-12/schema"
    • Changeddatacore_date1 field changed
      • changedInput schema / $schema
        Previous value: -"http://json-schema.org/draft-07/schema#"New value: +"https://json-schema.org/draft/2020-12/schema"
    • Changeddatacore_ingest1 field changed
      • changedInput schema / $schema
        Previous value: -"http://json-schema.org/draft-07/schema#"New value: +"https://json-schema.org/draft/2020-12/schema"
    • Changeddatacore_search1 field changed
      • changedInput schema / $schema
        Previous value: -"http://json-schema.org/draft-07/schema#"New value: +"https://json-schema.org/draft/2020-12/schema"
    • Changeddatacore_status1 field changed
      • changedInput schema / $schema
        Previous value: -"http://json-schema.org/draft-07/schema#"New value: +"https://json-schema.org/draft/2020-12/schema"
  2. 5 tool updatesv1.6.0
    • First observeddatacore_capture
    • First observeddatacore_date
    • First observeddatacore_ingest
    • First observeddatacore_search
    • First observeddatacore_status
Deploy Server
One endpoint for every MCP toolRayrun runs every MCP server behind one URL and holds the credentials. Each call is decided per client, tool, and argument, then recorded.ray.runAd

TDQS

A3.7/5.0

Scored across 5 tools

Disambiguation5/5

Each tool has a clearly distinct purpose: capture for journal entries, ingest for knowledge notes, search for querying, status for system health, and date for date operations. No overlap between any of the tools, so an agent can reliably select the correct one.

Naming Consistency5/5

All tools follow the same datacore_ prefix followed by a simple, consistent verb: capture, search, ingest, status, date. The naming pattern is uniform and predictable across the entire set.

Tool Count5/5

With 5 tools, the server is well-scoped for a knowledge management utility. Each tool earns its place, covering core operations plus a necessary date utility, and there is no bloat or redundancy.

Completeness4/5

The server covers the primary lifecycle of creating and searching notes, with status and date utilities supporting the workflow. Minor gaps include lack of update/delete operations for notes, but the core functionality is solid and agents can work around these limitations.

Maintenance

ActivityActive
ResponsivenessWithin a week

Resources

Related MCP Connectors

Related MCP Servers