database MCP Server
Exasol MCP server. Provides knowledge about the Exasol database to an LLM through the Model Context Protocol.
Discovered via github-seeds:mcp-hot and last synced 1d ago.
1. Add to claude_desktop_config.json
{
"mcpServers": {
"mcp-server": {
"command": "docker",
"args": [
"run",
"--rm",
"-i",
"ghcr.io/blockscout/mcp-server:latest"
]
}
}
}Config file location: ~/Library/Application Support/Claude/claude_desktop_config.json (macOS) / %APPDATA%\Claude\claude_desktop_config.json (Windows)
Aerospace engineers and systems architects who need to integrate Claude AI into their workflows benefit most from this MCP server. Software developers building aerospace applications can leverage it to add intelligent context management and reasoning capabilities. Teams developing AI-assisted tools for aircraft design, simulation, or mission planning will find it particularly valuable.
Feed aerospace design specifications and requirements into Claude through the MCP server, allowing the AI to maintain context across multiple design iterations and provide informed feedback on structural or aerodynamic trade-offs.
Use the server to pass flight parameters, constraints, and mission objectives to Claude, enabling it to help analyze and optimize flight plans while maintaining consistency across multiple planning scenarios.
Integrate aerospace technical documents and specifications through the MCP server so Claude can review, validate, and suggest improvements while maintaining full context of complex system requirements.
Build aerospace applications that communicate with Claude via this MCP server, enabling features like intelligent assistant functionality, real-time recommendations, and context-aware analysis within your existing tools.
This server implements the Model Context Protocol, enabling standardized communication between aerospace applications and Claude. It allows you to pass domain-specific context to Claude while maintaining protocol compliance, but the actual aerospace domain logic and capabilities depend on how you structure your inputs and prompts.
The server implements the standard MCP specification, so integration follows MCP patterns. You'll configure it as a data source in your Claude-compatible application or client, then route aerospace-related requests through it to ensure proper context management.
The server provides MCP protocol implementation for context management. Real-time data handling depends on your implementation—you can pipe streaming data through the server, but you'll need to handle the aerospace-specific processing and formatting on your end.
As an MCP server, it requires a compatible MCP host application and appropriate network configuration. Specific deployment requirements depend on your infrastructure, but you'll need Claude API access and a system capable of running the server process alongside your aerospace applications.
Run docker container
Comma-separated list of allowed tools
Solution
sources for cross-source comparison |
Pre-flight prompt safety evaluation against workspace policies
Read Claude Code project memory files (index or specific file)
Full details for an MCP server package: readme, tools, versions, reviews
**Live** at `mcp.thinkneo.ai`
SOC2, GDPR, HIPAA compliance readiness and governance score
URL
Cost savings report: original vs actual cost, breakdown by task type
AI Trust Score (0-100) across 6 dimensions with shareable badge
Set a pre-AI baseline metric for comparison
Verify a claim by checking evidence (HTTP status, file exists, DB row, etc.)
Verify if a model, provider, or action is allowed by governance policies
Write/update project memory files
Publish an MCP server: validates endpoint, runs security scan, stores entry
AI cost breakdown by provider, model, team, and time period
Real-time health and latency of 7 AI providers
Search the MCP Marketplace by keyword, category, rating, or verified status
Start an observability session for an agent
Log a risk that was avoided by AI intervention
Compare model performance across tasks
Log an agent-to-agent interaction with cost, outcome, and latency
Budget utilization, enforcement status, and projected overage
Book a demo with the ThinkNEO team
Rate and review an MCP server (1-5 stars, one per user per package)
Value
Active alerts and incidents for a workspace
Installation config for Claude Desktop, Cursor, Windsurf, or custom clients
Auth
API key usage stats: calls today/week/month, top tools, estimated cost
Calculate the cost of a specific decision
Dashboard of all claims, verification rates, and trust metrics
Prompt safety check: injection patterns (10 rules) + PII (credit card with Luhn, CPF with checksum, SSN, email, phone, API keys)
Find the cheapest model meeting a quality threshold across 17+ models from 9 providers
Simulate how much an organization would save with Smart Router
Log a tool call, model call, decision, or error event
End a trace session with final status
Aggregated observability metrics for a workspace
+--------+---------+
ROI report for a specific agent
Auto-generate an A2A Agent Card from the MCP tools/list
List all current MCP-A2A tool/skill mappings
Description
Get income statements for a company.
Get the current / latest price of a crypto currency.
Get news for a company.
Get balance sheets for a company.
Search restaurants in a city, with cuisine/price/feature filters
Monuments, museums, parks near a restaurant (Wikidata)
Today's time-based discount slots for a restaurant
Paginated reviews with star and date filters
Reserve a seat at a Flow-discounted slot
Full profile for a restaurant by slug
Restaurants within walking distance of a famous landmark
API URL
Restrict to read-only tools only
Need to Run Manually?
Configure MCP in app settings
Backend extracting fields/pages.
The sender cancelled the request.
int
yes
Ready for signing setup.
Processing failed.
int
string
Required by the MCP spec on every request to a Streamable HTTP server.
File still being received.
Fully signed; signed PDF + certificate page available.
object
string
at least one
string
string \
New color or `null` to clear.
string
File received, awaiting metadata extraction.
A signer rejected the document.
string
int
string
string
int
Accepted values
array
string
bool
string
yes
Assignment created; waiting on signers.
string
Meaning
Final signing artifact is being generated.
string
string
string
string
string
string \
no
Purpose
conditional
Assignment expired without completion.
string
string
string (ISO 8601)
string
string
string
bool
string
array
string
array
Type
string
Aggregate counts by status and form type
Forms
List payers across filings
Filing detail with payer summary and e-file status
All supported filing types, e-file availability, and filing system (IRIS/AIR/BSO/PDR)
List filings with filters (tax year, form type, status). Includes `filingSystem` field.
Payer details for a filing
Form metadata (type, status, dates)
Active
Active
E-file request/response timeline
E-file errors with codes and messages
— live pricing snapshot: plan tiers, feature matrix, add-ons, and the per-chain dedicated-node catalog
— send a message to Chainstack (sales, support, or general questions)
— top up a testnet address from the Chainstack faucet (12 networks across EVM, Solana, and TON)
— search Chainstack documentation
— check platform status and incidents
— fetch full doc page content
Gets all available crypto tickers.
Get the current / latest price of a company.
Gets historical stock prices for a company.
Gets historical prices for a crypto currency.
Gets historical prices for a crypto currency.
What it does
Balance + recent ledger entries
`https://api.ainfera.ai`
Top up an Agent's Wallet
—
Register a new Agent, returns ID + AgentCard
Most recent AuditEvents
`stdio`
List Agents for the authenticated Tenant
Verify a JWS AgentCard
Step-by-step guide for first-time Ainfera users
Make an Inference through L2 Routing
Notes
Mint a fresh JWS-signed AgentCard
Fetch an Agent by ID
Verify hash chain integrity
Read the organization account status.
Remove a collaborator from a project.
Update the contents of a file.
Update the name or note of a backtest.
Authorize an external connection with a live brokerage or data provider.
Stop a live algorithm.
Delete the Object Store file of a specific organization and key.
Description
Purpose
Create a new project in your default organization.
Lock a project so you can edit it.
Apply a patch (unified diff) to a file in a project.
Delete a backtest from a project.
Create a live algorithm.
Liquidate and stop a live algorithm.
Returns a list of LEAN versions with basic information for each version.
Update a project's name or description.
Asynchronously create a compile job request for a project.
Read the results of a backtest.
Read an optimization.
Read a chart from a live algorithm.
Upload files to the Object Store.
Show additional context and suggestions for error messages.
`string`
List the details of all projects.
Create a new backtest request and get the backtest Id.
Create an optimization with the specified parameters.
List all your past and current live trading deployments.
Broadcast a live command to all live algorithms in an organization.
Show the code completion for a specific text input.
`string` *optional*
`string` *optional*
List the details of a project or a set of recent projects.
Delete a project.
Add a collaborator to a project.
List all collaborators on a project.
Update collaborator information in a project.
Read the available and selected nodes of a project.
`boolean`
`integer`
`string`
`boolean`
`boolean` *optional*
`string`
`` *optional*
`object`
`string`
List forms in a filing
Returns remaining credits and current plan.
Transcript from a TikTok URL (`tiktok.com/@user/video/...` or `vm.tiktok.com`).
Transcript from an Instagram URL (`/reel/`, `/p/`, `/tv/`).
Transcript from a YouTube URL (watch, youtu.be, or Shorts).
Transcript from a Facebook URL (`facebook.com/watch`, `fb.watch`, Reels).
Lists recent transcripts saved to your account, with previews.
Create a governance policy for a workspace
Generate a benchmark report with recommendations
Set communication policies between agents (rate limits, allowed actions)
Register a verifiable claim: what an agent did, with evidence type
Receive an A2A task, find the matching MCP tool, execute it, return A2A response
Per minute
2 prompts with completions
Every response includes `X-RateLimit-Limit`, `X-RateLimit-Remaining`, `X-RateLimit-Reset`. Exceeded limits return `429` with `Retry-After`. See [docs/features/rate-limiting.md](docs/features/rate-limiting.md). --- ## Observability Integrations Native OpenTelemetry support — traces and metrics for every MCP tool call and A2A skill invocation. Enable with environment variables (disabled by default, zero overhead when off): ```bash OTEL_ENABLED=true OTEL_SERVICE_NAME=thinkneo-mcp-gateway OTEL_EXPORTER_OTLP_ENDPOINT=http://your-collector:4317 ``` **Metrics exported:** - `thinkneo.tool_calls_total` — Counter per tool + status (ok/error) - `thinkneo.tool_duration_seconds` — Histogram per tool - `thinkneo.active_requests` — Gauge of concurrent requests **Vendor guides:** - [Datadog](docs/integrations/datadog.md) - [Grafana Tempo + Prometheus](docs/integrations/grafana-tempo.md) - [Honeycomb](docs/integrations/honeycomb.md) --- ## Authentication - **API Key**: Prefix `tnk_`, provided via `Authorization: Bearer <key>` header - **OAuth 2.1**: Full authorization code flow supported - Public tools (safety checks, provider status, marketplace search) work without authentication ## Self-Signup Auto-provision a free API key (500 calls/month): ``` POST https://mcp.thinkneo.ai/mcp/signup ``` No credit card required. All 62 tools available on the free tier. ## Self-Hosted Deployment Run ThinkNEO gateway on your own infrastructure: ```bash git clone https://github.com/thinkneo-ai/mcp-server.git cd mcp-server/deploy cp .env.example .env # edit with your API key docker compose up -d # gateway + PostgreSQL ``` Gateway at `http://localhost:8081/mcp`. See [self-hosted quickstart](docs/self-hosted/quickstart.md). --- ## Enterprise Features - Per-minute rate limiting with configurable thresholds - IP allowlist per API key - Granular scopes (`read`, `execute`, custom) - API key rotation with revocation log - Enterprise tier with unlimited calls - OAuth 2.1 with client credentials flow - Dedicated support and SLA Contact: [[email protected]](mailto:[email protected]) --- ## Reliability Tested weekly against 5 failure scenarios using [toxiproxy](https://github.com/Shopify/toxiproxy) + [WireMock](https://wiremock.org/): 1. **PostgreSQL down** — public tools work, DB tools degrade gracefully, auto-recovery 2. **Redis down** — rate limiting fails open, no crash, auto-recovery 3. **Provider timeout** — 30s latency injected, gateway timeouts within 15s 4. **Provider 429** — handles rate limits, no infinite loop, per-key isolation 5. **High latency** — 5s latency, concurrent requests complete, P95 < 2s for pure tools See [chaos report](docs/reliability/chaos-report.md) and [runbook](docs/reliability/runbook.md). --- ## Security & Compliance - **Vulnerability reporting**: [SECURITY.md](SECURITY.md) — [email protected], 48h response - **Threat model**: [THREAT_MODEL.md](THREAT_MODEL.md) — STRIDE analysis - **Compliance posture**: [docs/compliance/current.md](docs/compliance/current.md) - **Compliance roadmap**: [docs/compliance/roadmap.md](docs/compliance/roadmap.md) — SOC 2 Type I in progress - **Data handling**: [docs/compliance/data-handling.md](docs/compliance/data-handling.md) - **security.txt**: `/.well-known/security.txt` ([RFC 9116](https://www.rfc-editor.org/rfc/rfc9116)) --- ## License [MIT](LICENSE) ## Links - **Website**: [thinkneo.ai](https://thinkneo.ai) - **Developer Docs**: [mcp.thinkneo.ai/mcp/docs](https://mcp.thinkneo.ai/mcp/docs) - **MCP Marketplace**: [mcp.thinkneo.ai/registry](https://mcp.thinkneo.ai/registry) - **Glama Listing**: [glama.ai/mcp/servers/ThinkneoAI/mcp-server](https://glama.ai/mcp/servers/ThinkneoAI/mcp-server) - **A2A Agent Card**: [mcp.thinkneo.ai/.well-known/agent.json](https://mcp.thinkneo.ai/.well-known/agent.json) - **Company**: ThinkNEO AI Technology Co. Ltd (Hong Kong)
completion/complete
Business impact report for a workspace
List all policies for a workspace
Explain why the Smart Router chose a specific model
Visualize agent-to-agent communication flows over a time period
List policy violations with severity and remediation
Translate an MCP tool call into an A2A task and dispatch it to an A2A agent
reserved
Retrieve full trace with all events for a session
Detect waste patterns: unused tools, redundant calls, idle agents
Evaluate an action against workspace policies
Generate an audit trail of all A2A interactions
1,000
Generate a compliance report for a framework
10
--- ## What This Is This repository contains the ThinkNEO Gateway, a dual-protocol server that exposes AI governance tools through both the **Model Context Protocol** (MCP, Anthropic 2024) and the **Agent-to-Agent Protocol** (A2A, Google/Linux Foundation 2025). The gateway implements a bidirectional **MCP-A2A bridge**: MCP clients can dispatch A2A tasks, and A2A agents can call MCP tools. Both protocols share the same governed runtime, meaning all traffic passes through the ThinkNEO Enterprise AI Control Plane for policy enforcement, audit logging, spend tracking, and compliance checks. The server is production-deployed as a single Docker container, serving MCP on `/mcp` and the A2A Agent Card at `/.well-known/agent.json`. All 62 MCP tools and 24 A2A skills run in the same process, share the same PostgreSQL database, and are wrapped with the same free-tier middleware (500 calls/month, auto-provisioned API keys). --- ## MCP Tools 62 tools across 12 categories. All tools accept JSON-RPC 2.0 via Streamable HTTP. ### Governance (6)
Public trust score badge lookup by token
Log an AI agent decision with type, rationale, and context
Get the full proof chain for a verified claim
List available compliance frameworks and their status
600
Get cash flow statements for a company.
`boolean`
`array` *optional*
`string`
`string`
`string`
`string`
`string`
`integer`
`number` *optional*
`string`
`object` *optional*
`integer`
`integer`
Add a file to a given project.
Read a chart from a backtest.
Update the name of an optimization.
Read out the portfolio state of a live algorithm.
Check the syntax of a code.
`string`
`object`
Read a compile packet job result.
Read a file from a project, or all files in the project if no file name is provided.
Update the name of a file.
Delete a file in a project.
List all the backtests for the project.
Read out the orders of a backtest.
Read out the insights of a backtest.
List all the optimizations for a project.
Abort an optimization.
Delete an optimization.
Read details of a live algorithm.
Get the logs of a live algorithm.
Read out the orders of a live algorithm.
Read out the insights of a live algorithm.
Send a command to a live trading algorithm.
Get Object Store properties of a specific organization and key.
Get the URL for downloading files from the Object Store.
List the Object Store files under a specific directory in an organization.
Run a backtest for a few seconds to initialize the algorithm and get inialization errors if any.
Update Python code to follow PEP8 style.
Returns the latest version of the QC MCP Server released.
`boolean`
`string`
`integer` *optional*
`integer` *optional*
`string` *optional*
`array` *optional*
`array`
`array`
`array`
`string`
`string`
`integer`
`string`
`integer`
`object`
`integer` *optional*
`number`
`status enum` *optional*
`string`
`string`
`string`
`string`
`string` *optional*
`string`
What to check
Comma-separated list of tools to exclude
Detects
Description
SQL injection, XSS, command injection, path traversal
Credential stuffing, brute force, token replay, privilege escalation
Rate abuse, resource exhaustion, API scraping
2 resources
logging/setLevel
Description
Header anomalies, spoofing, missing security headers
Path probing, tool enumeration, method probing, fingerprinting
Description
Status
72 tools, full annotations
`string`
Update the active state of the given nodes to true.
Estimate the execution time of an optimization with the specified parameters.
Create a job to download files from the Object Store and then read the job Id.
Search for content in QuantConnect.
Returns the version of the QC MCP Server that's running.
Tools
`compare_traces`, `compare_traces_over_time`
MCP transport mode
`get_replay_command`
`resolve_call_ids`
`create_test_group`, `list_test_groups`, `list_test_group_classes`, `list_test_group_methods`, `build_test_payload`, `delete_test_group`, `set_test_group_enabled`, `regenerate_test`, `get_test_results`
Host for HTTP mode
Port for HTTP mode
Purpose
Responsibility
Exposes MCP tools and adds comparison, normalization, and formatting logic
Purpose
Default token when a tool call does not pass `mcp_token`
Base BitDive Monitoring API URL
Disable TLS certificate verification when set to `true`
Typed HTTP client from an OpenAPI 3.x document, passed inline or by URL — 10 frameworks
Typed client from a GraphQL SDL schema, optionally with reusable named fragments — 10 frameworks
Typed client from Protocol Buffers (`.proto`) definitions — 10 frameworks
Typed model classes from SQL DDL (`CREATE TABLE`), parsed dialect-agnostically — 11 languages
Arbitrary type graphs (classes, interfaces, enums, generics) from one structured spec, with imports and cross-references resolved — 11 languages
Runs a `generate_code` spec from disk, so multi-file architectures stay version-controlled and context usage drops to a file path
Primes the agent before its first generation: patterns, examples, and language-specific rules
Type
`string`
`string`
`array`
What it does
Use this skill when designing database schemas, optimizing queries, creating indexes, planning migrations, or choosing between database technologies. Triggers on schema design, normalization, indexing strategies, query optimization, EXPLAIN plans, migrations, partitioning, replication, connection pooling, and any task requiring database architecture or performance decisions.
Database Workflow Bundle workflow skill. Use this skill when the user needs Database development and operations workflow covering SQL, NoSQL, database design, migrations, optimization, and data engineering and the operator should preserve the upstream workflow, copied support files, and provenance before merging or handing off.
database-admin workflow skill. Use this skill when the user needs Expert database administrator specializing in modern cloud databases, automation, and reliability engineering and the operator should preserve the upstream workflow, copied support files, and provenance before merging or handing off.
AppDeploy Skill workflow skill. Use this skill when the user needs Deploy web apps with backend APIs, database, and file storage. Use when the user asks to deploy or publish a website or web app and wants a public URL. Uses HTTP API via curl and the operator should preserve the upstream workflow, copied support files, and provenance before merging or handing off.
open-source agentic AI data assistant for the next generation of AI + Data products.
MCP Toolbox for Databases is an open source MCP server for databases.
Data Agent Ready Warehouse : One for Analytics, Search, AI, Python Sandbox. — rebuilt from scratch. Unified architecture on your S3.
Second Brain is an agentic framework that acts as an operating system, using local file intelligence, workflow automation, and LLMs to complete tasks and communicate over multiple modalities and messaging platforms.
Learn how to use the database-engineering Claude skill. Complete guide with installation instructions and examples.
Learn how to use the azure-postgres-ts-v2 Claude skill. Complete guide with installation instructions and examples.
Learn how to use the database-architect-v3 Claude skill. Complete guide with installation instructions and examples.