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/alexk-dev/golemcore-bot/claude-mdgit clone --depth 1 https://github.com/alexk-dev/golemcore-botWrote 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/alexk-dev/golemcore-bot/claude-md)<a href="https://agentmods.dev/instructions/alexk-dev/golemcore-bot/claude-md"><img src="https://agentmods.dev/badge/instructions/alexk-dev/golemcore-bot/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 | $0.03570 | $0.03570 |
| Opus 5 | $0.01785 | $0.01785 |
| Sonnet 5 | $0.00714 | $0.00714 |
| Haiku 4.5 | $0.00357 | $0.00357 |
Grade A, and why
golemcore-bot 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 — 362 lines — stays where its author put it; the contents beside it link to each section on GitHub.
CLAUDE.md — GolemCore Bot
Build & Test
./mvnw clean package -DskipTests # build all modules
./mvnw test # run tests
./mvnw clean verify -P strict # full check (tests + PMD + SpotBugs)
To run the app locally, configure provider API keys in preferences/runtime-config.json (recommended: via the dashboard at /dashboard).
Common env vars (Spring) for local/Docker:
STORAGE_PATH(workspace base path)TOOLS_WORKSPACE(sandbox for filesystem/shell tools)
Modules
golemcore-bot-contracts— shared contracts: ports, shared domain models, contract-style component interfaces, lightweight shared views.golemcore-bot-app— runnable Spring Boot application and runtime implementation.
Rule: new shared contracts must go into golemcore-bot-contracts, not into the app module.
Use golemcore-bot-contracts for:
- port interfaces
- shared boundary/domain models used across modules
- contract-style component interfaces
- lightweight shared views
Do not put controllers, adapters, Spring services, or tool implementations into golemcore-bot-contracts.
Architecture
Spring Boot 4.0.5, Java 25, Hexagonal Architecture (Ports & Adapters).
Package Structure
me.golemcore.bot
├── adapter/inbound/ Telegram (TelegramAdapter, TelegramVoiceHandler), CommandRouter, Webhook (WebhookController)
├── adapter/outbound/ LLM, storage, browser, MCP, voice (ElevenLabsAdapter)
├── domain/component/ Interfaces: ToolComponent, SkillComponent, etc.
├── domain/loop/ AgentLoop, AgentContextHolder
├── domain/model/ AgentContext, AgentSession, Message, Skill, ToolDefinition, ContextAttributes, etc.
├── domain/service/ SessionService, SkillService, CompactionService, VoiceResponseHandler, etc.
├── domain/system/ Ordered pipeline systems (see below)
├── port/inbound/ ChannelPort, CommandPort
├── port/outbound/ LlmPort, StoragePort, BrowserPort, VoicePort, McpPort, SessionPort, RagPort, ConfirmationPort
├── security/ InjectionGuard, InputSanitizer, AllowlistValidator
└── tools/ FileSystemTool, ShellTool, BrowserTool, TierTool, VoiceResponseTool, etc.
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 · 362 lines · 3,570 tokens per session scan A 7dd33802e517
golemcore-bot CLAUDE.md is an instructions file published in the GitHub repository alexk-dev/golemcore-bot (40 stars, last pushed 1mo ago), licensed Apache-2.0. It adds 3,570 tokens to every session, about $0.0178 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
LIA-Assistant AGENTS.md
AGENTS.md instructions for jgouviergmail/LIA-Assistant, covering agents.md, project overview, build & development commands, full setup (backend + frontend + git hooks) and stop.
LIA-Assistant CLAUDE.md
Claude Code instructions for jgouviergmail/LIA-Assistant, covering claude.md, project overview, build & development commands, full setup (backend + frontend + git hooks) and stop.
LobsterAI AGENTS.md
AGENTS.md instructions for netease-youdao/LobsterAI, covering agents.md, instruction scope, project snapshot, cowork vs openclaw and commands.
ai CLAUDE.md
Claude Code instructions for TanStack/ai, covering claude.md, repository overview, package manager & tooling, common commands and testing.
ai AGENTS.md
AGENTS.md instructions for TanStack/ai, covering agents.md, docs skill (mandatory for claude, grok, and codex), bugfix pr skill (mandatory for claude, grok, and codex), contributing guide (mandatory for issues and prs) and ponytail skill (mandatory for claude, grok, and codex).
orbit CLAUDE.md
Claude Code instructions for schmitech/orbit, covering claude.md, project overview, commands, setup and running the server.