Semantic intelligence layer for SQL Server databases, exposed via MCP. A database understanding engine for AI agents (Claude Code CLI, Codex CLI, Claude Desktop, etc.).
Wellknown found it in public sources; nobody has proven control of it yet. Claiming takes one click if the repository is under your GitHub account, or a small file on your domain otherwise. Verified owners get the badge, 15-minute checks, status alerts, edits that outrank crawled data, and a ranking boost.
Agents can do it too: POST https://wellknown.network/api/v1/claims with {"agent":"sqlserver-semantic-mcp","method":"well_known_file"} — machine-readable steps at claim.json, guide at /docs/claim.
Everything here was measured by our prober or read from a registry. Nothing is self-reported.
Attributed to the source that supplied each field. Treated as claims, not facts.
# sqlserver-semantic-mcp [](LICENSE) [](https://www.python.org/) [](https://modelcontextprotocol.io) [](pyproject.toml) [](README.zh-TW.md) > **Semantic intelligence layer for SQL Server databases, exposed via MCP.** > Not a SQL executor — a database understanding engine for AI agents. AI agents don't need raw `execute_sql`. They need to understand schema structure, relationships, object dependencies, and — most importantly — to operate inside a safety boundary that an operator can define. `sqlserver-semantic-mcp` provides all of this through 12 MCP tools, 1 concrete MCP resource, and 5 MCP resource templates, backed by a two-tier SQLite cache for speed and a JSON-based policy system for safety. --- ## Quick Start Pick the path that matches your client. All paths use [`uvx`](https://docs.astral.sh/uv/) — no `git clone`, no virtualenv, no manual install. `uvx` downloads and runs the package on demand and caches it for next time. > **Prerequisite:** Install [uv](https://docs.astral.sh/uv/getting-started/installation/) once (`curl -LsSf https://astral.sh/uv/install.sh | sh`). Python 3.11+ is fetched automatically by `uv` if needed. > **Replace** `localhost` / `YourDatabase` / `sa` / `YourPassword` in every example with your real SQL Server credentials. ### 🤖 Claude Code CLI One command registers the server. `uvx` resolves and caches `sqlserver-semantic-mcp` on first use: ```bash claude mcp add sqlserver-semantic -- uvx sqlserver-semantic-mcp \ -e SEMANTIC_MCP_MSSQL_SERVER=localhost \ -e SEMANTIC_MCP_MSSQL_DATABASE=YourDatabase \ -e SEMANTIC_MCP_MSSQL_USER=sa \ -e SEMANTIC_MCP_MSSQL_PASSWORD=YourPassword ``` Or commit the config to your repo as `.mcp.j…
Mapped onto the structured taxonomy from declared text and observed tool names. Confidence shown for derived entries.
Every source is kept verbatim. Field changes are logged as events.