Portward
Related Servers
Alternatives to Portward
No user-submitted related servers found.
Related Servers
- AlicenseNot gradedqualityCmaintenanceRead-only MCP server for SQLite databases, enabling AI assistants to safely query and inspect database schemas without write access.MIT
- AlicenseNot gradedqualityDmaintenanceRead-only MCP server for SQL databases (SQLite/PostgreSQL) that enables listing tables, describing schemas, and executing SELECT queries with safety guardrails.MIT
- FlicenseNot gradedqualityDmaintenanceA read-only MCP server that enables AI agents to explore database schemas and execute safe queries on PostgreSQL and MySQL.-
- AlicenseNot gradedqualityDmaintenanceA read-only MCP server that enables LLMs to safely explore and query any SQLite database via natural language. It exposes tools for listing tables, describing schemas, and executing SELECT/WITH queries with built-in safety guards like write prevention and row limits.MIT
- AlicenseAqualityCmaintenanceRead-only MCP server that lets AI agents safely query SQLite, PostgreSQL, and MySQL/MariaDB. Enforces read-only transactions with column masking, row caps, query timeouts, EXPLAIN-based cost rejection, and rate limiting.734 npm2MIT
- AlicenseNot gradedqualityCmaintenanceAn MCP server that provides safe, read-only SQL access for AI agents to query databases (PostgreSQL, MySQL, SQLite) with schema awareness and guardrails.5 npmMIT
TDQS
Scored across 77 tools
The core data path (query, rerun, run_saved, tweak, compare, kept_rows) is documented carefully and mostly separable, but with 77 tools several boundaries blur: use_block_rules vs unapply_profile, kept_rows vs panel_rows vs more_rows, assumptions vs define_assumption vs draft_definition, and share vs share_sign_in vs share_confirm. A third of the tools are explicitly app-only (never seen by the model), which trims the practical surface but the remaining set still has clusters where an agent must re-read descriptions to choose.
Everything is snake_case, which is good, but verb conventions vary widely: prefixed verbs (list_sources, get_selection, set_note, remove_source) sit alongside bare verbs (query, tweak, compare, block, share, export) and bare nouns (history, lineage, assumptions, shares, panel_rows, context_drafts). Readable, but not a single predictable pattern an agent can rely on.
77 tools is far past what an agent-facing surface should carry; even discounting the app-only panel/secret tools, the model still faces 60+ options across querying, sharing, profiles, context drafting, sync, account and routing. It behaves like several servers glued together, and the volume itself drives misselection.
Coverage of the domain is genuinely deep: sources (add/test/describe/remove), querying and history, saved pulls, sharing with expiry/withdrawal, profiles and grants, block rules, context drafts and definitions, sync, account, and export/proof reporting. Only minor gaps remain, e.g. no way to delete or prune history entries or a pull's children.