MCP ConfigsMay 13, 2026·2 min read

agent-tool — Encoding-Safe MCP Tools for Agents

agent-tool is an MCP tool server with encoding-aware, indentation-safe file tools and an `install` command for Claude Code, Codex, Cursor, and more.

MCP Hub
MCP Hub · Community
Agent ready

Safe staging for this asset

This asset is staged first. The copied prompt tells the agent to inspect the staged files and ask before activating scripts, MCP config, or global config.

Stage only · 17/100Policy: stage
Agent surface
Any MCP/CLI agent
Kind
Mcp Config
Install
Stage only
Trust
Trust: Established
Entrypoint
Asset
Safe staging command
npx -y tokrepo@latest install ecce188d-d7be-5c01-995d-256baccaf7ed --target codex

Stages files first; activation requires review of the staged README and plan.

Intro

agent-tool is an MCP tool server with encoding-aware, indentation-safe file tools and an install command for Claude Code, Codex, Cursor, and more.

Best for: repos with mixed encodings or strict indentation rules (tabs/.editorconfig)

Works with: any MCP-capable agent; auto-install supports Claude Code/Codex/Cursor/Cline (per README)

Setup time: 5-12 minutes

Key facts (verified)

  • GitHub: 17 stars · 2 forks · pushed 2026-05-05.
  • License: MIT · owner avatar + repo URL verified via GitHub API.
  • README-verified entrypoint: agent-tool install --safe-approve.

Main

  • Use agent-tool install --safe-approve to register with detected agents while limiting auto-approval to local file tools (README lists the safety tiers).

  • When encoding matters (EUC-KR / Shift-JIS / GB18030), set a fallback encoding via --fallback-encoding or env var as shown in README.

  • Treat it as a tool unifier: README frames it as bundling many tools into one install so agents stop reconfiguring per-project utilities.

Source-backed notes

  • README lists auto-install commands: agent-tool install, --safe-approve, and --no-auto-approve.
  • README includes manual MCP config examples for Claude Code/Cursor/Codex (mcpServers JSON or [mcp_servers.*] TOML).
  • README explains it addresses indentation and encoding corruption issues in built-in file tools.

FAQ

  • Is safe mode recommended?: README recommends --safe-approve when you only want local file tools auto-approved.
  • Can I configure it manually?: Yes — README shows JSON/TOML examples for different clients.
  • How do I handle non-UTF-8 repos?: Set --fallback-encoding or AGENT_TOOL_FALLBACK_ENCODING as shown in README.
🙏

Source & Thanks

Source: https://github.com/knewstimek/agent-tool > License: MIT > GitHub stars: 17 · forks: 2

Discussion

Sign in to join the discussion.
No comments yet. Be the first to share your thoughts.

Related Assets

Postgres MCP Pro — Index Tuning + Safe SQL Tools

Postgres MCP Pro is an MCP server for PostgreSQL that runs safe SQL, explains plans, and recommends indexes so agents can tune databases faster.

MCP Configs
MCP Hub

Notte — Browser Automation MCP for AI Agents

MCP server that turns web browsers into AI agent tools. Notte provides structured browser actions like click, type, navigate, and extract for LLM-driven automation.

MCP Configs
MCP Hub

WhatsApp MCP Server — Chat with WhatsApp via AI Agents

MCP server connecting Claude and AI agents to your personal WhatsApp. Search contacts, read messages, send replies and media via natural language.

MCP Configs
MCP Hub

Apify MCP Server — 8,000+ Web Scrapers for Agents

Apify MCP Server connects agents to Apify Actors via a hosted endpoint (mcp.apify.com) or local run, turning thousands of web scrapers into callable tools.

MCP Configs
MCP Hub
◈Home🔍Search👤Me