Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add instructions/div197/bob-gemini-free/agents-mdgit clone --depth 1 https://github.com/div197/BOB-Gemini-FreeWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/instructions/div197/bob-gemini-free/agents-md)<a href="https://agentmods.dev/instructions/div197/bob-gemini-free/agents-md"><img src="https://agentmods.dev/badge/instructions/div197/bob-gemini-free/agents-md.svg" alt="Measured on agentmods" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.02608 | $0.02608 |
| Opus 5 | $0.01304 | $0.01304 |
| Sonnet 5 | $0.00522 | $0.00522 |
| Haiku 4.5 | $0.00261 | $0.00261 |
Grade B, and why
BOB-Gemini-Free AGENTS.md scanned grade B with 1 finding against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 5d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Asks for rootmediumPrivilege escalation
A mod that escalates privileges can change anything on the machine, not only the project.
- Save via `./bob-gemini-free --setup-cookie` or write directly to `./cookie.txt` (`chmod 600`). How it starts
The opening of the file, as written. The whole thing — 197 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AGENTS.md — AI Agent Guidelines for BOB Gemini Free
Welcome, AI Agent! This document provides technical architecture, build instructions, directory layouts, and execution workflows for automated agents (Cursor, Windsurf, Claude Code, Codex CLI, Aider, OpenHands, Devin, Roo Code) interacting with the BOB Gemini Free codebase.
1. Project Overview
BOB Gemini Free (Break Ordinary Boundaries) by ABCsteps (abcsteps.com) and Divyanshu Singh Chouhan (@div197) is a high-performance, single-binary local AI gateway written in Go.
It translates three major protocol standards into Google's internal web RPC protocol (batchexecute / streamGenerateContent):
- OpenAI Standard:
POST /v1/chat/completions,POST /v1/responses(withoutput_text),POST /v1/tokens/count,GET /v1/models,GET /v1/models/{model},POST /v1/images/generations - Anthropic Standard:
POST /v1/messages(Claude Code CLI with complete SSE stream lifecycle and prompt caching counters) - Google Gemini Standard:
GET /v1beta/models,POST /v1beta/models/{target}(includinggenerateContent,streamGenerateContent, and:countTokens) - Embedded Go Library:
pkg/gateway(import "github.com/div197/bob-gemini-free/pkg/gateway")
2. Directory & Package Structure
.
├── main.go # Entrypoint & CLI flag routing
├── go.mod / go.sum # Go module definition (Go 1.26.6 in this snapshot)
├── Makefile # Multi-arch compilation & test commands
├── install.sh / install.ps1 # Cross-platform automated setup scripts
├── test-kit.sh / test-kit.ps1 # Automated diagnostic runners
├── pkg/
│ └── gateway/ # Exported embedded Go library (pkg/gateway)
└── internal/
├── browser/ # 1-click native interactive login window (CDP WebSocket)
├── config/ # Configuration loader & cookie auto-discovery
├── diag/ # 15-point diagnostic test kit, bench runner & e2e tests
├── format/ # Protocol translation (OpenAI, Anthropic, Google, Images, Citations, Tokens)
├── gemini/ # Upstream client, auth, cookie pool load-balancer, parser
├── models/ # Model registry & alias catalog (Imagen 3, Nano Banana 2/Pro)
├── multimodal/ # Image compression, Scotty upload, token cache, dynamic refresh
└── server/ # HTTP mux, middleware, SSE streamer, handlers
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 5d ago First seen · 197 lines · 2,608 tokens per session scan B 0558abc6f510
BOB-Gemini-Free AGENTS.md is an instructions file published in the GitHub repository div197/BOB-Gemini-Free (3 stars, last pushed today), licensed MIT. It adds 2,608 tokens to every session, about $0.0130 per session on Opus 5. A static security scan graded it B with 1 finding (asks for root). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other instructions, from other repositories
adk-go CLAUDE.md
Claude Code instructions for google/adk-go: See AGENTS.md for project context, commands, and contribution guidelines for AI coding agents.
aisix CLAUDE.md
Claude Code instructions for api7/aisix, covering claude.md, 1. think before coding, 2. simplicity first, 3. surgical changes and 4. goal-driven execution.
race-condition copilot-instructions.md
Copilot instructions for GoogleCloudPlatform/race-condition, covering race condition -- ai assistant instructions, quick orientation, project conventions and detailed task skills.
race-condition AGENTS.md
AGENTS.md instructions for GoogleCloudPlatform/race-condition, covering race condition -- ai assistant instructions, quick orientation, project conventions and detailed task skills.
race-condition CLAUDE.md
Claude Code instructions for GoogleCloudPlatform/race-condition, covering race condition -- ai assistant instructions, quick orientation, project conventions and detailed task skills.
race-condition GEMINI.md
Gemini CLI instructions for GoogleCloudPlatform/race-condition, covering race condition -- ai assistant instructions, quick orientation, project conventions and detailed task skills.