Summary
If your AI coding agent is burning tokens reading generated files, build artifacts, and node_modules every session, this intercepts MCP file operations and strips framework noise before it hits the model. It registers optimized_read_file, optimized_list_files, optimized_grep, and optimized_diagnostics tools that apply framework-aware ignore rules (Python, JS/TS included, with paid packs for React, Flutter, Go, and others). Works with Claude Code, Cursor, Cline, Continue, and anything else speaking stdio MCP. The core is MIT licensed with automatic detection via pyproject.toml or package.json. Pro tier adds LSP integration for read_symbol and history compaction. Install globally with npm, point your agent's config at the ai-optimizer command, and it claims 60-80% token reduction on typical sessions.
Fastest browser for AI agents to run web automation tasks, always free.
Download Free life-time →
Notes, actions and memory. Without a meeting bot. First month 100% off.
Download for free →
belt cli automatically finds the best tools and skills for your agent. image, video, music, tts...
one prompt install →
Connect your Claude agent to live crypto prices and trading routes via 1inch
Get the MCP →
Block distracting apps from your iPhone permanently without a 3rd party app. Free and open source.
Block now (100% free) →AI Token Optimizer
An MCP server that cuts AI coding-agent token usage by 60–80% via framework-aware context optimization.
AI coding agents read whole files when they need 15 lines. They load *.g.dart,
R.java, node_modules/, and dist/ into context — then do it again next turn.
AI Token Optimizer is a Model Context Protocol
server that intercepts those tool calls and strips framework noise before it
reaches the model.
Works with anything that speaks MCP: Claude Code, Cursor, Cline, Continue, Zed, JetBrains AI, GitHub Copilot agent mode, Antigravity, Windsurf, OpenAI Codex CLI.
Quickstart
npm install -g @ai-optimizer/core
Then point your agent at the ai-optimizer stdio server. For Claude Code, add to
~/.claude/settings.json:
{
"mcpServers": {
"ai-optimizer": {
"command": "ai-optimizer"
}
}
}
Or scaffold the config and preview your savings automatically:
npx @ai-optimizer/init
# "we'd skip 47 files, ~14K tokens/session, ~$0.42/session at Sonnet pricing"
Full per-agent setup (Cursor, others), LSP install, and verification steps live in
packages/core/README.md.
What it does
It registers tools the agent uses instead of its built-in file/dir/error reads:
| Tool | What it saves |
|---|---|
optimized_read_file | Framework-aware skip + optional line slice |
optimized_list_files | Walks the tree applying ignore globs (+ .gitignore) |
optimized_grep | Content search scoped by ignore filters |
optimized_diagnostics | Spawns the right LSP, returns errors + a narrow code window |
read_symbol | LSP workspace/symbol slice instead of a full-file read (Pro/LSP) |
strip_bash_noise | Strips ANSI, npm/maven/gradle progress, JVM warnings |
optimizer_status | Detected frameworks, active packs, cumulative tokens saved |
When a project type is detected (e.g. Python via pyproject.toml), the matching
pack's ignore rules activate automatically. .gitignore is respected on top.
Tiers
| Free (MIT core) | Pro ($9/mo) | Team ($29/seat) | |
|---|---|---|---|
| Framework packs | Python, JS/TS | + React, Flutter, Java, Kotlin, Go, Rust, C#, Swift, Ruby, Elixir, PHP, Solidity | all Pro packs |
| History compaction | — | ✓ | ✓ |
Scheduler (claude -p cron) | — | ✓ | + work-stealing across providers |
| Audit log + secret redaction + policy | — | — | ✓ |
See PRICING.md. Behavior-changing or data-sharing features
(scheduler, work-stealing, telemetry) are off by default and require explicit
opt-in. Display-only features (dollar-value savings) default on.
Monorepo layout
| Package | Purpose | Published |
|---|---|---|
packages/core | MCP server + free packs | npm (MIT) |
packages/init | npx @ai-optimizer/init config + savings preview | npm |
packages/sdk | definePack plugin SDK | npm |
packages/context-mode-import | Import .contextmode configs | npm |
packages/pro | Pro/Team packs + scheduler + audit | private |
apps/landing, apps/dashboard, apps/vscode-extension, apps/license-server | Site, local dashboard, IDE widget, license issuance | private |
tools/pack-bench | Pack quality benchmark harness | private |
Architecture details: ARCHITECTURE.md.
Development
npm install # at repo root (workspaces)
npm run build # build all packages
npm test # run all test suites
Security
Local-only by design — license validation is the only network call. Report
vulnerabilities per SECURITY.md.
License
Core engine is MIT — see LICENSE. Pro/Team packs are proprietary.
Featured
Fastest browser for AI agents to run web automation tasks, always free.
Download Free life-time →Notes, actions and memory. Without a meeting bot. First month 100% off.
Download for free →belt cli automatically finds the best tools and skills for your agent. image, video, music, tts...
one prompt install →Connect your Claude agent to live crypto prices and trading routes via 1inch
Get the MCP →Block distracting apps from your iPhone permanently without a 3rd party app. Free and open source.
Block now (100% free) →Configuration
AI_OPTIMIZER_LICENSEsecretEd25519 JWT license key. Unlocks Pro/Team packs; free tier needs no key.
AI_OPTIMIZER_LSPSet to 0 to disable the LSP bridge (optimized_diagnostics).
AI_OPTIMIZER_TELEMETRYSet to 'on' to enable anonymous opt-in telemetry. Off by default.
Categories
Registryactive
Package@ai-optimizer/core
TransportSTDIO
AuthRequired
UpdatedJun 6, 2026
