Mantora
About
Mantora is a local-first MCP observer: a lightweight UI + proxy for inspecting LLM data access (sessions, tool calls, results) with protective defaults.
Details
- Author
- josephwibowo
- Categories
- Database, AI
Jump to
Setup
Install Mantora in your MCP client (Claude Desktop, Cursor, Windsurf, and others).
Repository: https://github.com/josephwibowo/mantora
Follow the installation instructions in the repository README, then restart your MCP client.
The open-source run log for MCP agents.Get a shareable report to verify and reproduce what the agent did —plus a protective mode for common dangerous SQL.
Mantora focuses on making agent behaviorreviewable and repeatable. It records the key evidence from each session and turns it into a Markdown report. Protective mode adds basic guardrails (block/warn rules) so you don’t learn about a bad query the hard way.
Capture the exact SQL and tool calls an agent makes for easy verification.
Basic guardrails that block or warn on destructive SQL patterns by default.
One-click to generate a Markdown summary for Pull Requests or Slack.
pipx install "mantora[duckdb,postgres]"
Configure your MCP client (e.g.,claude_desktop_config.jsonor Cursor Settings) to run the Mantora proxy. No target configuration is needed here!
{ "mcpServers": { "mantora": { "command": "mantora", "args": ["mcp"] } } }
Start the dashboard to manage your connections and view logs in real-time:
Note: For advanced configuration (safety policy, storage paths), you can use amantora.tomlfile. SeeConfigurationfor details.
sequenceDiagram participant Agent as Claude/Cursor participant Mantora as Mantora 🛡️ participant DB as Postgres/DuckDB Agent->>Mantora: "SELECT FROM users" Note over Mantora: 1. Check Safety Policy Mantora->>DB: Execute Query DB-->>Mantora: Return Rows Note over Mantora: 2. Log Report (Async) Mantora-->>Agent: Relay Result
Don't paste messy screenshots. In the Session Summary, click"Copy for GitHub".
- Result:A collapsible<details>block containing the SQL, execution time, and safety checks. Perfect for Pull Requests.
If an agent tries toDROP TABLE users, Mantora intercepts it.
- The Agent sees:"Error: Action blocked by safety policy."
- You see:A red entry in the Mantora UI with the exact SQL that was blocked.
- Export JSON:Get the raw trace data.
- Download Markdown:Get a readable log of the entire conversation.
Mantora is designed to beread-only safeout of the box.
Want to change this? SeeConfiguration.
- DuckDB(Local files, MotherDuck)
- Postgres(Direct connection, Supabase, Neon)
- BigQuery, Snowflake, Databricks(Coming Soon)
All data is stored locally. SeePrivacyfor details.*
- Configuration Guide- Flags, ports, and policy overrides.
- Architecture- How the proxy internals work.
- Contributing- Build setup and testing.
- Open an Issue
- Star this repo⭐️ if you find it useful!
Hydrolix time-series datalake integration providing schema exploration and query capabilities to LLM-based workflows.
614 million contacts, live SMTP-verified. Invalid emails cost you $0. Connect Argorant with any Agent through CLI, MCP or API and build verified lists in Claude, GPT or Hermes.
Chinese writing toolbox for Claude Code: text analysis, outline generation, GB/T 7714 references,character frequency
A lightweight MCP server for connecting to Apache Doris and other MySQL-compatible databases, providing tools and prompts for LLM applications.
C# SQL Agent MCP server featuring raw SQL input, strict AST validation, and an embedded Admin UI. Eliminates LLM hallucinations and security risks across 6 major databases.
Build and access a personal LLM knowledge base from your editor or client without any infrastructure setup.
Biologically-inspired memory for Claude. Hebbian plasticity, dream cycles, and temporal decay for persistent knowledge synthesis.
Access the Quran API from alquran.cloud to retrieve accurate Quranic text and reduce LLM hallucinations.
An MCP server for local Tabular Models like PowerBI. It allows LLM clients to debug, analyze, and compose DAX queries by connecting to a local Tabular model instance.
Interact with Verodat AI Ready Data platform
Sign in to leave a review
Use Google, GitHub, or an email account so ratings stay tied to real people.
No reviews posted yet.
