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/livelykernel/lively4-core/claude-mdgit clone --depth 1 https://github.com/LivelyKernel/lively4-coreWhat 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 | $0.05158 | $0.05158 |
| Opus 5 | $0.02579 | $0.02579 |
| Sonnet 5 | $0.01032 | $0.01032 |
| Haiku 4.5 | $0.00516 | $0.00516 |
Grade A, and why
lively4-core 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 3d 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 — 490 lines — stays where its author put it; the contents beside it link to each section on GitHub.
CLAUDE.md
This file provides guidance to Claude Code (claude.ai/code) when working with code in this repository.
IMPORTANT: Object-Oriented Architecture
This is NOT a typical functional JavaScript project!
Lively4 follows object-oriented principles with class hierarchies and inheritance:
- Use inheritance properly: If functionality needs to be shared, put it in the base class
- Single source of truth: Don't duplicate methods across subclasses - use the parent class
- Polymorphism: Subclasses inherit and can override parent methods when needed
- Class hierarchy matters:
LivelyChat→OpenaiRealtimeChat,LivelyOpencode,LivelyAiWorkspace
Example: Database Write Guards
- ❌ WRONG: Add
canWriteToDatabase()to each component separately - ✅ RIGHT: Add
canWriteToDatabase()toLivelyChatbase class, all subclasses inherit it
Example: Composition Pattern - Propagating State
When a parent component contains child components, state changes must be explicitly propagated:
// ❌ WRONG: Only set state on parent
enableReplay() {
this._replayMode = true; // Only affects parent, children still write to DB!
}
// ✅ RIGHT: Propagate state to composed children
enableReplay() {
this._replayMode = true;
// CRITICAL: Propagate to child components
this.realtimeComponent._replayMode = true;
this.opencodeComponent._replayMode = true;
}
disableReplay() {
this._replayMode = false;
// CRITICAL: Clear from child components
if (this.realtimeComponent) this.realtimeComponent._replayMode = false;
if (this.opencodeComponent) this.opencodeComponent._replayMode = false;
}
Key principle: Child components don't automatically inherit instance variables from their container. State must be explicitly synchronized in composition relationships.
When you find yourself duplicating code across components that share a base class, STOP and move it to the parent class instead.
AI Collaboration Experiment
A significant part of this AI collaboration is an experiment to teach Claude Code how to develop in Lively4. By working together on real development tasks, we are:
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.
- 3d ago First seen · 490 lines · 5,158 tokens per session scan A 0576978bb253
lively4-core CLAUDE.md is an instructions file published in the GitHub repository LivelyKernel/lively4-core (88 stars, last pushed 10d ago), licensed MIT. It adds 5,158 tokens to every session, about $0.0258 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
FlyEnv AGENTS.md
AGENTS.md instructions for xpf0000/FlyEnv, covering flyenv - ai agent development guide, project overview, technology stack, core technologies and additional libraries.
coder AGENTS.md
AGENTS.md instructions for coder/coder, covering coder development guidelines, task-specific guidance, workflow, essential commands and repository guardrails.
flox AGENTS.md
AGENTS.md instructions for flox/flox, covering claude.md, project overview, development setup, already in dev shell (innixshell is set) — run directly and not in dev shell — wrap with nix develop -c.
coder CLAUDE.md
Claude Code instructions for coder/coder, a project described as: Secure environments for developers and their agents.
flox CLAUDE.md
Claude Code instructions for flox/flox, a project described as: The Deterministic Foundation for your SDLC.
crystal CLAUDE.md
Instructions for stravu/crystal, covering crystal - multi-session claude code manager, project overview, references, implementation status: ✅ complete and ✅ implemented features.