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/stewart86/bobby/claude-mdgit clone --depth 1 https://github.com/Stewart86/bobbyWrote 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/stewart86/bobby/claude-md)<a href="https://agentmods.dev/instructions/stewart86/bobby/claude-md"><img src="https://agentmods.dev/badge/instructions/stewart86/bobby/claude-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 | $0.01487 | $0.01487 |
| Opus 5 | $0.00744 | $0.00744 |
| Sonnet 5 | $0.00297 | $0.00297 |
| Haiku 4.5 | $0.00149 | $0.00149 |
Grade A, and why
bobby CLAUDE.md scanned grade A with 0 findings 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 4d 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.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 166 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Bobby - Discord AI Assistant Bot
Overview
Bobby is a Discord chatbot that helps answer questions about your codebase, file bugs, and translate business requirements into technical requirements. Bobby leverages Claude Code to understand your codebase and provide intelligent responses in a self-hosted, privacy-first environment.
Core Architecture
Main Components
- index.js - Main Node.js application using Bun runtime
- Discord.js - Handles Discord bot interactions and thread management
- Claude Code CLI - Powers AI analysis of codebases with streaming responses
- GitHub CLI - Creates issues automatically when bugs are detected
- Docker Container - Self-contained environment with all dependencies
Environment Variables Required
DISCORD_TOKEN=your_discord_bot_token
ANTHROPIC_API_KEY=your_anthropic_api_key
GH_TOKEN=your_github_personal_access_token
GITHUB_REPO=owner/repo-name
ALLOWED_DISCORD_SERVERS=server_id1,server_id2 (optional)
Docker Architecture
- Base Image:
oven/bun:latest - Dependencies: Git, GitHub CLI v2.61.0, Claude Code CLI
- Working Directory:
/app - Repository Clone:
/app/repo(target repository) - Data Storage:
/app/data(Claude Code sessions and configs) - Entrypoint:
entrypoint.sh(handles authentication and setup)
Thread-Based Session Management
Bobby uses Discord threads for session management with streaming responses:
- New Conversations: Mention Bobby in any channel to create a new thread with a new Claude Code session
- Follow-ups: Simply type in the thread (no need to mention Bobby) to continue the same Claude Code session
- Thread Naming: Threads are automatically named as
Bobby - Title - session-idwhere:Titleis a 3-5 word summary generated by Claude (hidden from user via regex)session-idis the Claude Code session identifier
- Streaming Responses: Real-time message streaming with JSON parsing
- Auto-Archive: Threads automatically archive after 24 hours of inactivity
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.
- 4d ago First seen · 166 lines · 1,487 tokens per session scan A 781418979e93
bobby CLAUDE.md is an instructions file published in the GitHub repository Stewart86/bobby (11 stars, last pushed 1y ago), licensed MIT. It adds 1,487 tokens to every session, about $0.0074 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other instructions, from other repositories
artbot AGENTS.md
Instructions for ArtBlocks/artbot, covering artbot — agent & maintainer guide, what this is, quick commands, two production flags (critical) and architecture (actual).
WhatsAppToDiscord AGENTS.md
AGENTS.md instructions for arespawn/WhatsAppToDiscord, covering agent notes (agents.md), project snapshot, fast start, where the source of truth lives and critical guardrails.
artbot CLAUDE.md
Instructions for ArtBlocks/artbot: Project memory for Claude Code lives in AGENTS.md (shared with Cursor and other agents). Operational checklists: docs/MAINTENANCE.md.
DiscordMCP CLAUDE.md
Claude Code instructions for Rastrian/DiscordMCP, covering discord mcp platform — claude project instructions, project identity, core framing, strict non-goals and product goal.
discord-mcp AGENTS.md
Instructions for sandraschi/discord-mcp, covering discord-mcp — agent guide, standards, key files and quick ref.
discord-mcp CLAUDE.md
Instructions for sandraschi/discord-mcp, covering discord-mcp — claude code guide, standards and key files.