A lightweight stdio server that hands your AI agent 27 curated operations LLMs routinely fumble: SHA-256 hashing, UUID generation, JWT decoding, regex matching, WCAG contrast checks, and bulk format/convert tasks like JSON-to-TypeScript or SQL pretty-printing. Runs locally via npx with zero telemetry. The tool set is deliberately narrow because every MCP tool costs context on each agent request, so this skips what models already do well and focuses on deterministic ops where guessing wastes tokens or produces wrong answers. Same engine that powers the consolelog.tools web app and CLI, so results are identical across surfaces. Works with Claude Code, Cursor, Windsurf, and other MCP clients.
io.github.infoinlet-marketplace/mcp-observability
betterdb-inc/monitor
com.mcparmory/datadog
thotischner/observability-mcp
io.github.tantiope/datadog-mcp
io.github.us-all/datadog