AgentStack
MCP unreviewed MIT Self-run

Toksave

mcp-agungprasastia-toksave · by agungprasastia

A zero-config CLI that securely installs and wires token-saving tools into AI coding agents.

No reviews yet
0 installs
13 views
0.0% view→install

Install

$ agentstack add mcp-agungprasastia-toksave

Open-source listing — not yet scanned by AgentStack. Follow the source repository for install instructions.

Security review

⚠ Flagged

1 finding(s); flagged for manual review. · v0.1.0 How review works →

  • Prompt-injection patterns
  • Secret / credential exfiltration
  • Dangerous shell & filesystem operations
  • Untrusted network calls
  • Known-malicious package signatures
  • high Pipes remote content directly into a shell (remote code execution).

What it can access

  • Network access Used
  • Filesystem access No
  • Shell / process execution No
  • Environment & secrets No
  • Dynamic code execution No

From automated source analysis of v0.1.0. “Used” means the capability is present in the source — more access means more to trust, not that it’s unsafe.

View the full security report →

Reliability & compatibility

Not yet reviewed
0 installs to date
no reviews yet
27d ago

Declared compatibility

Claude CodeClaude DesktopCursorWindsurf

Compatibility is declared by the source manifest. End-to-end runtime verification is coming — see below.

Preview Execution monitoring

We're building live execution health for every listing: tool-call success rate, median latency, uptime, and last-checked timestamps — measured, not self-reported. It isn't live yet, so we don't show numbers we can't stand behind.

How agent discovery & health will work →
Are you the author of Toksave? Claim this listing to set pricing, connect Stripe payouts, and keep 70% of every sale.
Sign up to claim

About

toksave

Zero-config token-saver for AI coding agents.

[](https://opensource.org/licenses/MIT) [](https://bun.sh) [](https://www.typescriptlang.org/) [](https://github.com/agungprasastia/toksave/actions)

Install and wire RTK, Caveman, CodeGraph, and Context-Mode into your AI agents — without hand-editing configs.

✨ Features

  • Zero-config: Run one command to equip all your agents.
  • Smart Health Checks: Built-in diagnostics verify installations and toksave doctor --fix can repair unhealthy tools.
  • Actionable Error Messages: When things fail, get clear explanations and remediation steps.
  • Download Resilience: Automatic retry with exponential backoff handles flaky networks.
  • Idempotent & Safe: Run multiple times without duplicating configurations.
  • Clean Uninstall: Tracks what it installed so it can cleanly revert changes.
  • Cross-platform: Ships as a standalone binary for macOS, Linux, and Windows.
  • Preview Changes: Use --dry-run to see what will happen before committing.

🤖 Supported Agents & Tooling

| Agent | MCP | Hooks | Caveman | Context-Mode Rules | |-------|:---:|:---:|:---:|:---:| | Claude Code | ✅ | RTK hook + instructions | ✅ SKILL.md | ✅ AGENTS.md | | OpenCode | ✅ | RTK plugin + instructions | ✅ AGENTS.md | ✅ AGENTS.md | | Codex | ✅ (TOML) | RTK hook + instructions | ✅ instructions.md | ✅ instructions.md | | Antigravity | ✅ (JSON) | RTK hook + instructions | ✅ SKILL.md | ✅ AGENTS.md |

📦 What Gets Installed

| Tool | Method | What It Does | |------|--------|-------------| | RTK | Prebuilt binary | CLI proxy that compresses tool output — 60-90% token savings | | Caveman | Markdown from GitHub | Communication mode that cuts LLM response tokens ~75% | | CodeGraph | npm install -g | Pre-indexed code knowledge graph — fewer tool calls | | Context-Mode| npm install -g | MCP server that sandboxes tool output — 98% context reduction |

🚀 Getting Started

Prerequisites

  • Node.js ≥ 22 (required for CodeGraph and Context-Mode)
  • At least one supported AI agent installed

Install

macOS / Linux:

curl -fsSL https://raw.githubusercontent.com/agungprasastia/toksave/main/scripts/install.sh | bash

Windows (PowerShell):

irm https://raw.githubusercontent.com/agungprasastia/toksave/main/scripts/install.ps1 | iex

💻 Usage

Quick Start

# Interactive setup — detect agents, install tools, wire everything
toksave

Advanced Commands

| Command | Description | |---------|-------------| | toksave | Install + wire all tools into detected agents | | toksave doctor | Read-only health check with diagnostics and repair suggestions | | toksave doctor --fix | Repair unhealthy tool installations, then report health after repair | | toksave update | Update all tools to latest versions | | toksave uninstall | Remove toksave wiring from agents | | toksave self-update | Update the toksave CLI itself |

Helpful Flags

| Flag | Description | |------|-------------| | -a, --agents | Target specific agents (e.g., claude,antigravity) | | -t, --tools | Target specific tools (e.g., rtk,caveman) | | -n, --dry-run | Preview changes without modifying anything | | -y, --yes | Skip prompts, auto-select detected agents (CI-friendly) | | -v, --verbose | Show detailed output |

🛠️ Development

Built with TypeScript + Bun. Compiles to a standalone binary via bun build --compile.

# Clone and install dependencies
git clone https://github.com/agungprasastia/toksave.git
cd toksave
bun install

# Development tasks
bun run src/index.ts      # Run CLI in dev mode
bun run typecheck         # Run TypeScript checks
bun test                  # Run unit tests
bun run lint              # Lint with Biome
bun run build             # Build local binary

# Build all cross-platform releases
bash scripts/build-release.sh

See [CHANGELOG.md](CHANGELOG.md) for release history and detailed changes.

📜 License

Licensed under the MIT License — [see LICENSE](LICENSE).

Source & license

This open-source MCP server is cataloged on AgentStack and links to its original source — we do not rehost the code.

Install and usage instructions live in the source repository linked above.

Reviews

No reviews yet — be the first.

Versions

  • v0.1.0 Imported from the upstream source.