Skip to content

Instantly share code, notes, and snippets.

View bgauryy's full-sized avatar

bgauryy

27 followers · 4 following
  • Tel Aviv
View GitHub Profile
@bgauryy
bgauryy / Claude_Prompt_Engineering_Cheatsheet.md
Created January 26, 2026 09:35

Claude Prompt Engineering Cheatsheet

Best practices for writing instructions that Claude reliably follows.

1. Command Strength Hierarchy

Use stronger language for critical instructions:

@bgauryy
bgauryy / clawdbot-security-audit.md
Last active January 25, 2026 12:08
Clawdbot Security Audit Report - A code-verified security assessment of the clawdbot WhatsApp AI assistant (v2026.1.24-0) identifying 18 security issues across critical, high, and medium severity levels. Key findings: (1) WhatsApp messages including phone numbers, GPS locations, and group contacts flow unsanitized to external AI providers (Anthr…

Clawdbot Code Analysis Report

@bgauryy
bgauryy / LOCAL_TOOLS_TEST_PLAN.md
Created January 1, 2026 20:41
Comprehensive test framework (380+ test cases) comparing Octocode MCP local tools against Claude Code and Cursor built-in tools across 9 test suites. Covers four tool categories: code search, directory listing, file reading, and file finding. Tests span 10 dimensions: context quality, efficiency, output quality, token safety, security, error han…

Local Tools Test Plan: AI Coding Assistants vs Octocode MCP

Comprehensive test plan comparing AI coding assistant internal tools with Octocode MCP local tools

Objective: Validate that Octocode local tools provide superior context, efficiency, output quality, token safety, and security compared to built-in tools in Claude Code and Cursor.

Tool Mapping Overview

@bgauryy
bgauryy / OCTOCODE_CURSOR_LOCAL_COMPARISON.md
Last active January 1, 2026 20:42
Head-to-head comparison of Octocode MCP local tools vs Cursor's built-in tools on the Linux kernel (100K+ files). Tests four tool categories—code search, directory listing, file reading, and file finding—using identical queries. Evaluates output structure, metadata richness, pagination, token efficiency, and actionable guidance. Result: Octocode…

🔬 ADVANCED DEPTH COMPARISON: Octocode MCP vs Cursor Internal Tools

Repository: Linux Kernel (100K+ Files)

🎯 Motivation

Cursor provides built-in local tools (grep, list_dir, read_file, glob_file_search) that AI agents use to explore and understand codebases. These tools are fast, simple, and familiar—but are they sufficient for large-scale code research?

@bgauryy
bgauryy / octocode_security.md
Created November 20, 2025 17:58
Octocode output doc for a security review doing on dzhng/claude-agent-server repo using review prompt

Security Audit Report: claude-agent-server

Repository: https://github.com/dzhng/claude-agent-server
Audit Date: November 20, 2025
Auditor: Security Review via Octocode
Severity Scale: Critical > High > Medium > Low

Executive Summary

@bgauryy
bgauryy / data-format-token-test.js
Created November 15, 2025 08:46
Measures token count using the cl100k_base tokenizer (used by GPT-4/Claude) for CSV, TOON, JSON
import { encode as toonEncoder } from '@byjohann/toon'
import { createByEncoderName } from '@microsoft/tiktokenizer'
import { stringify } from 'csv-stringify/sync'
import { writeFileSync, mkdirSync } from 'fs'
import { join } from 'path'
// Helper function to convert data to CSV format
function dataToCSV(data) {
// Handle arrays of objects (most common CSV case)
if (Array.isArray(data)) {
@bgauryy
bgauryy / AGENTIC_AI_LEARNING_PATH.md
Created November 9, 2025 08:27
Octocode AI Research Summary On - github. com/panaversity/learn-agentic-ai
@bgauryy
bgauryy / React_Advanced_Concepts_Research.md
Created November 9, 2025 06:15
React Advanced Concepts Used WIth Octocode MCP
@bgauryy
bgauryy / octocode_plan.ms
Created October 30, 2025 17:29
Octocode Plan For Agentic Application
# AI Agent Chat Application - Implementation Plan
## Research Summary
### Repositories Analyzed
- **langchain-ai/langgraphjs-gen-ui-examples** (347⭐) - LangGraph.js agent examples
- **assistant-ui/assistant-ui** (6,896⭐) - TypeScript/React AI chat UI library
- **FlowiseAI/Flowise** (46,168⭐) - Visual AI agent builder with LangChain
- **CopilotKit/CopilotKit** (24,685⭐) - React UI + infrastructure for AI agents
@bgauryy
bgauryy / script.js
Created October 27, 2025 14:56
TOON vs MINIFIED JSON TOKENIZATION - regular object
import { encode as toonEncoder } from '@byjohann/toon'
import { createByEncoderName } from '@microsoft/tiktokenizer'
import { writeFileSync, mkdirSync } from 'fs'
import { join } from 'path'
async function main() {
const data = {};
// Generate large dataset programmatically
const emails = ['john.doe', 'jane.smith', 'bob.wilson', 'alice.johnson', 'charlie.brown', 'david.miller', 'emma.davis', 'frank.garcia', 'grace.martinez', 'henry.rodriguez'];