Agent Booster

Name: Agent Booster
Author: sseshachala

by sseshachalaversion 0.6.4i★ –iRuns on your computer

Registry nameio.github.sseshachala/agent-booster

Install

Overview

Description from the maker (original English)

Cut AI agent token costs 5-15x — routes only relevant code symbols instead of full files.

PopNx MCP’s three pillars

These three are never merged into one score: opinions, measurements and popularity are different things.

User voiceiNo reviews yetReviews open together with accounts in a later step
Our basic checksi1 of 1 passedChecked from public information; code not scanned yet
UsageiNot measured yet★ –

What you can do with it

We have not written a plain-language summary for this one yet. Below are the facts we read from the registry.

Connection: Runs on your computer
Transport: stdio
Category: Other

Install

Pick the AI app you use, then copy the settings.

Before you install

Basic checks: 1 of 1 passed
Runs on your computer with your user’s permissions (files, network, apps).
The registry lists no key to fill in.

You need

PyPIPython’s uv installed (for the uvx command)

Which app do you use?

ClaudeClaude CodeCursorVS CodeChatGPT

Pick your app above. We remember it, so next time the button says “Install in …”.

Open Claude Desktop → Settings → Developer → Edit Config, and paste this into claude_desktop_config.json

{
  "mcpServers": {
    "agent-booster": {
      "command": "uvx",
      "args": [
        "agent-booster"
      ]
    }
  }
}

Run this in your terminal

claude mcp add agent-booster -- uvx agent-booster

Open in Cursor

Or copy the settings yourself

Paste into .cursor/mcp.json in your project, or ~/.cursor/mcp.json

{
  "mcpServers": {
    "agent-booster": {
      "command": "uvx",
      "args": [
        "agent-booster"
      ]
    }
  }
}

Your browser asks to open the app, then the app shows the settings and asks you to confirm. Key values are left empty for you to fill in.

Open in VS Code

Or copy the settings yourself

Paste into .vscode/mcp.json in your project

{
  "servers": {
    "agent-booster": {
      "type": "stdio",
      "command": "uvx",
      "args": [
        "agent-booster"
      ]
    }
  }
}

Your browser asks to open the app, then the app shows the settings and asks you to confirm. Key values are left empty for you to fill in.

This one runs on your computer, and ChatGPT can only connect to servers on the internet. Pick another app.

Your choice stays on this device in a cookie; no account needed.

These settings are generated from the maker’s registry entry. If the README says otherwise, follow the README.

What it can do

Runs on your computer; the tool list appears in your AI app after you install it.

Basic checksi

We check information anyone can see; we have not read the code. Checks that do not apply are left out of the count. · Checked Nov 7, 2026

✓ Pass
Maker matches the repo
The registry name was verified with the GitHub account sseshachala, which owns the repo
… Not checked yet
Still maintained
We have not read this repo from GitHub yet
… Not checked yet
Has a licence
We have not read this repo from GitHub yet
– Not applicable
Endpoint responds
This one runs on your computer; there is no internet endpoint
– Not applicable
Sign-in with OAuth
Runs on your computer and uses local settings instead
… Not checked yet
No hidden instructions in tool descriptions
Runs on your computer; we can only read tool lists from internet endpoints

Compared with others in Otheri

More stars than 0% of the other 32673 servers in this category
Published to the registry more recently than 14% of the other 32673 servers in this category
Basic checks equal to or better than 100% of the other 32673 servers in this category

Compared with every active server in this category in the registry. Stars count only when the publisher owns the repo.

Common questions

Is Agent Booster safe?

It passed 1 of 1 applicable basic checks. We have not read the code. Before adding a key or token, see what access the server asks for and use a key with only the permissions you need.

What keys or codes does it need?

The registry does not list a required key, though some servers ask you to sign in on first use.

Does it run on my computer or in the cloud?

It runs on your computer. Your AI app starts it when needed, and data does not pass through the maker’s servers (unless it calls an online service itself).

Is it still maintained?

The GitHub repo last changed –, and the latest version (0.6.4) reached the registry 4 months ago.

Which apps can use it?

Apps that support MCP, such as Claude Desktop, Claude Code, Cursor and VS Code. You also need Python (uv) on your computer.

What can I use instead?

In the same category: Spec Driven Development, Remote Desktop Commander, Systemdox

Alternatives

Servers in the same category from other makers, sorted by stars.

Reading a card:★ GitHub starsiThree pillars: reviews · checks · usageiUpdated = latest version published to the registryi

Spec Driven Development

by formulahendry

MCP Server that facilitates spec-driven development workflows, not just Vibe Coding.

Runs on your computer★ 440MITUpdated 14 months ago

Install

Remote Desktop Commander

by desktop-commander

Remote Desktop Commander: use files, terminal and processes on your own computers from any AI app.

Runs in the cloud★ 211Updated 2 days ago

Install

Systemdox

by puglieseweb

Your architecture in your agents' context: checks, specs, ADRs and docs over MCP, at write time.

Runs on your computerUpdated today

Install

README from the maker

Open the README

Copied from the maker’s repo as written. PopNx MCP has not reviewed or endorsed it. HTML, images and scripts are removed. Links lead off this site.

This repository is archived. Agent Booster now lives in sseshachala/conductai → tools/booster. Install from PyPI (pip install agent-booster); file issues and PRs on conductai.

Agent Booster

Cut AI coding agent token costs 5–15x by routing only the code that matters.

Instead of sending full source files to the model on every read, Agent Booster builds a symbol index of your codebase and returns only the functions and classes relevant to the current task.

How it works

Without Booster                    With Booster

───────────────── ──────────────────────────── Read executor.py smart_read executor.py + task → 1,881 lines (~6k tokens) → 3 matching functions (~200 tokens)

Five layers work together:

Layer	What it does
Symbol index	tree-sitter parses every `.py`, `.ts`, `.tsx`, `.js`, `.jsx` file, extracts functions/classes into SQLite
Vector embeddings	sentence-transformers encodes each symbol for semantic search
Smart read	given a file + task, runs per-file vector search and returns only matching symbol line ranges
MCP server	exposes four tools over stdio — any MCP-compatible agent can call them
Platform init	one command writes the right config for Claude Code, Cursor, Windsurf, or Codex

Installation

Requires Python 3.10+.

pip install agent-booster          # symbol index + MCP tools

pip install agent-booster[embed] # + semantic vector search (recommended)

Quickstart

Step 1 — Wire up your AI tool

booster init claude     # Claude Code

booster init cursor # Cursor booster init windsurf # Windsurf booster init codex # OpenAI Codex CLI booster init all # all four

Each command shows exactly what files will change, asks for confirmation, then writes them. Fully reversible:

booster remove claude   # or cursor, windsurf, codex, all

Example — wiring up OpenAI Codex (booster init codex → booster start → booster gain):

Run once per project, not once per session.

The hooks live in .claude/settings.json and .mcp.json at the project root.

Every Claude Code session opened in this directory — any terminal window, same machine — picks them up automatically.

No per-session setup needed.

Step 2 — Index your codebase

Run once from your project root. Re-run after large refactors.

booster index && booster embed

Indexed 260 files, 1800 symbols.

Built embeddings for 1800 symbols.

The index is stored at .booster/ (gitignored). The index is shared across all sessions.

Step 3 — Restart your AI tool

The agent-booster MCP server is now available in every session. Claude Code (and other tools) will use smart_read and search_context automatically on indexed files.

Step 4 — Track savings

booster gain

CLI reference

`booster init`

Writes the MCP server config and rules file for the target platform. Asks for confirmation before making any changes.

booster init claude     # .mcp.json + CLAUDE.md + .claude/settings.json hook

booster init cursor # .cursor/mcp.json + .cursorrules booster init windsurf # ~/.windsurf/mcp.json + .windsurfrules booster init codex # ~/.codex/config.json + AGENTS.md booster init all # all four

booster init claude --yes # skip confirmation (CI/scripts)

`booster remove`

Cleanly undoes everything init wrote. No residue.

booster remove claude

booster remove cursor booster remove windsurf booster remove codex booster remove all

`booster index`

Scans all .py, .ts, .tsx, .js, .jsx files from the current directory. Extracts functions, classes, methods, and interfaces into .booster/symbols.db. Skips node_modules, .venv, __pycache__, .git, .booster, .next, dist, build.

booster index

Indexed 260 files, 1800 symbols.

`booster embed`

Builds sentence-transformer vector embeddings for all indexed symbols. Required for semantic search_context calls. Uses all-MiniLM-L6-v2 (local, no data leaves your machine).

booster embed

Built embeddings for 1800 symbols.

`booster search " "`

Keyword search across all indexed symbols. Returns file path, line number, kind, name, and signature.

booster search "guard install"

apps/web/src/app/settings/modules/page.tsx:99 function handleInstall function handleInstall() {

`booster route " "`

Recommends haiku, sonnet, or opus based on task complexity — keyword signals, file count, and symbol count.

booster route "fix the guard SSE role check"

haiku (narrow task — 0 symbol(s) in 1 file)

booster route "refactor the entire runtime compiler and DSL layer"

opus (matches complexity keywords)

`booster serve`

Starts the MCP server over stdio. Called automatically by Claude Code, Cursor, Windsurf, and Codex when configured via booster init. You rarely need to run this directly.

`booster gain`

Shows token savings from past smart_read calls — total reads, tokens served vs. tokens saved, savings rate, and top files by savings.

Agent Booster — Token Savings Report

───────────────────────────────────── Active days: 3 Total reads: 47 Tokens served: 12,400 Tokens saved: 89,200 Savings rate: 88%

Top files by savings: executor.py 18,400 tokens saved (12 reads) guard.py 14,200 tokens saved (8 reads)

MCP tools

Once booster serve is running, these four tools are available to the agent:

`get_symbols(file)`

Returns all indexed symbols for a file — name, kind, line range, and signature. No file read required.

Input:  { "file": "apps/api/app/runtime/executor.py" }

Output: function run_workflow (lines 42-89): def run_workflow(...) class WorkflowState (lines 91-140): class WorkflowState:

`search_context(task)`

Semantic vector search across all symbols in the index. Returns top 10 matches by cosine similarity. Falls back to keyword search if embeddings haven't been built.

Input:  { "task": "guard install uninstall" }

Output: apps/web/src/app/settings/modules/page.tsx:99 function handleInstall — function handleInstall() { apps/web/src/components/guard/GuardNav.tsx:13 function GuardNav — function GuardNav() { ...

`smart_read(file, task)`

Runs per-file vector search and returns only the source lines for matching symbols, with a header showing name and line range. If no symbols match, returns an explicit message so the model knows to fall back to a full Read rather than silently receiving the whole file.

Input:  { "file": "apps/api/app/runtime/executor.py", "task": "execute output block" }

Output: # function _execute_output (lines 1165-1210) def _execute_output(block, state, credentials, ...): ...

`route_model(task, files?)`

Recommends haiku, sonnet, or opus based on task complexity signals. If files is omitted, auto-detects via search_context.

Input:  { "task": "fix the login redirect bug" }

Output: { "model": "haiku", "reason": "narrow task — 1 symbol in 1 file" }

What `booster init claude` writes

File	Change
`.mcp.json`	Adds `agent-booster` to `mcpServers`
`CLAUDE.md`	Appends booster usage rules block (sentinel markers)
`.claude/settings.json`	Wires `PreToolUse` (Read + Grep) and `UserPromptSubmit` hooks
`.claude/hooks/booster-gate.py`	Blocks `Read` on indexed files — forces `smart_read`
`.claude/hooks/booster-grep-nudge.py`	Nudges semantic `Grep` patterns toward `search_context`
`.claude/hooks/booster-route.py`	Auto route_model — recommends haiku/sonnet/opus on every user turn

booster remove claude removes all six, cleanly. No residue.

Where it fits

Agent Booster is the third layer in a three-layer token reduction stack:

Layer 3 — Agent Booster     AST+semantic routing, smart file reads

Layer 2 — RTK Token compression on CLI/git/build output Layer 1 — Prompt caching Stable context reuse (native to Claude Code + API)

Each layer is independent and addable separately.

Common questions

Does it work if I open a second terminal? Yes. The hooks are wired into .claude/settings.json and .mcp.json in your project root — not inside a specific terminal session. Any new Claude Code window you open in this project picks them up automatically.

Do I need to run `booster init` again after restarting my machine? No. booster init writes files to disk once. They persist across restarts and across sessions. The only time you re-run it is if you delete those files or switch to a new project.

Do teammates need to run `booster init` too? Yes — once per developer, once per project. If you commit .mcp.json and .claude/settings.json to the repo, teammates get the hooks when they clone. They still need to run pip install agent-booster[embed] and booster index && booster embed locally, since the symbol index is gitignored.

Can I undo it completely? Yes. booster remove claude removes all six files and cleans the settings entries. No residue.

Claude Desktop

Add to ~/Library/Application Support/Claude/claude_desktop_config.json (macOS) or %APPDATA%\Claude\claude_desktop_config.json (Windows):

"mcpServers": { "agent-booster": { "command": "booster", "args": ["serve"] } } }

Run booster index && booster embed once from your project root before starting Claude Desktop. No env vars required.

Project layout

tools/booster/

├── README.md ├── pyproject.toml └── booster/ ├── cli.py # click commands: index, embed, search, serve, init, remove, route, gain ├── indexer.py # tree-sitter parser + SQLite symbol store + vector search ├── retriever.py # smart_read: per-file vector search → relevant line slice ├── mcp_server.py # MCP server: get_symbols, search_context, smart_read, route_model └── stats.py # token savings tracking (booster gain)

Agent Booster

Overview

PopNx MCP’s three pillars

What you can do with it

Install

Before you install

What it can do

Basic checksi

Maker matches the repo

Still maintained

Has a licence

Endpoint responds

Sign-in with OAuth

No hidden instructions in tool descriptions

Compared with others in Otheri

Common questions

Alternatives

Spec Driven Development

Remote Desktop Commander

Systemdox

README from the maker

Agent Booster

How it works

Installation

Quickstart

Indexed 260 files, 1800 symbols.

Built embeddings for 1800 symbols.

CLI reference

booster init

booster remove

booster index

Indexed 260 files, 1800 symbols.

booster embed

Built embeddings for 1800 symbols.

booster search " "

apps/web/src/app/settings/modules/page.tsx:99 function handleInstall function handleInstall() {

booster route " "

haiku (narrow task — 0 symbol(s) in 1 file)

opus (matches complexity keywords)

booster serve

booster gain

MCP tools

get_symbols(file)

search_context(task)

smart_read(file, task)

route_model(task, files?)

What booster init claude writes

Where it fits

Common questions

Claude Desktop

Project layout

`booster init`

`booster remove`

`booster index`

`booster embed`

`booster search " "`

`booster route " "`

`booster serve`

`booster gain`

`get_symbols(file)`

`search_context(task)`

`smart_read(file, task)`

`route_model(task, files?)`

What `booster init claude` writes