TokRepo MCP Server

The GitHub for AI assets · humans instruct, agents operate

TokRepo - Where AI Experience Lives

Infrastructure for the AI era. Manage your skills, MCPs, prompts and scripts in one place — humans confirm intent, agents install or push through TokRepo tools.

Browse trending

Connect TokRepo through MCP or CLI so any capable agent can search, evaluate, plan, install, and push assets through TokRepo.

Live agent usage · last 14 days See full funnel →

88

Agent events

9

Discoveries

22

Successful installs

2

Cross-session handoffs

Agent-native surface

Agents call TokRepo tools directly

Search, detail, verify, install-plan, and push flows are structured contracts for MCP/CLI agents; the web page only hands off the contract.

Search

tokrepo_search(query)

Detail

tokrepo_detail(uuid)

Verify

tokrepo_verify(uuid)

Plan

tokrepo_install_plan(uuid)

★ Editor's spotlight

Standalone gems — hand-picked, refreshed every 4 hours

The arsenal

[

Database MCP Stack — Query Safely Across Systems

8 assets in this pack

Ten verified database MCP servers and memory layers to query Postgres, MongoDB, ClickHouse, BigQuery, Trino, and dbt assets with guardrails.

• ▸Postgres MCP Pro — Index Tuning + Safe SQL Tools
• ▸pgEdge Postgres MCP — Read-Only by Default + Web UI
• ▸MongoDB MCP Server — npx + Read-Only Safety by Default

$ tokrepo install pack/database-mcp-stack-2026

](https://tokrepo.com/en/packs/database-mcp-stack-2026)[

MCP Security & Ops Toolkit — Scan Before You Connect

8 assets in this pack

Eight tools to run MCP gateways, aggregate servers, convert docs to Markdown, and scan MCP/agent stacks for security risks before production.

• ▸ContextForge — MCP Registry + Proxy Gateway
• ▸MetaMCP — Aggregate Many MCP Servers as One
• ▸Exa MCP Server — Remote Search Tools for Agents

$ tokrepo install pack/mcp-security-ops-toolkit-2026

](https://tokrepo.com/en/packs/mcp-security-ops-toolkit-2026)[

Agent Ops Toolchain (Observability + Security)

8 assets in this pack

A promoted pack for operating agents safely: scoped auth CLIs, eBPF traces, prompt/MCP security scans, a gateway layer, and orchestrated multi-agent workflows.

• ▸PentestAgent — MCP-Ready AI Pentesting Agent
• ▸Memori — Agent-Native Memory Infrastructure
• ▸AxonHub — Open-Source AI Gateway Platform

$ tokrepo install pack/agent-ops-observability-security-toolchain-2026-05

](https://tokrepo.com/en/packs/agent-ops-observability-security-toolchain-2026-05)[

MCP Builders Stack — Context, Sandbox, Automation

8 assets in this pack

Promoted pack for agent builders: cut context with code-review-graph, build servers with FastMCP, automate via Playwright/Chart/Excel MCPs, manage specs with Spec Workflow, and isolate runs in AIO Sandbox.

• ▸code-review-graph — MCP Context for Smarter Reviews
• ▸FastMCP — TypeScript Framework for MCP Servers
• ▸Playwright MCP — Browser Automation for Agents

$ tokrepo install pack/mcp-builders-context-sandbox-stack-2026-05

](https://tokrepo.com/en/packs/mcp-builders-context-sandbox-stack-2026-05)[

Agent Ops Guardrails Toolchain (2026)

8 assets in this pack

Eight GitHub-verified picks to run coding agents safely: macOS sandboxing, multi-channel gateways, kernel-tracing MCP, safety SDKs, Claude Code governance, and JSON repair.

• ▸json_repair — Fix Malformed LLM JSON in Python
• ▸aidevops — AI DevOps Framework with Agent Playbooks
• ▸Agent Safehouse — Sandbox macOS Coding Agents

$ tokrepo install pack/agent-ops-guardrails-toolchain-2026

](https://tokrepo.com/en/packs/agent-ops-guardrails-toolchain-2026)[

Structured Outputs Stack — Instructor, BAML, Guidance

6 assets in this pack

Six GitHub-verified picks to make agents reliable: typed extraction (Instructor), schema-first prompts (BAML), constrained generation (Guidance/Outlines), flow testing (PromptFlow), and tracing (Weave).

$ tokrepo install pack/structured-outputs-2026

](https://tokrepo.com/en/packs/structured-outputs-2026)