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/wildchildforlife/lantern/agents-mdgit clone --depth 1 https://github.com/WildChildForLife/lanternWrote 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/wildchildforlife/lantern/agents-md)<a href="https://agentmods.dev/instructions/wildchildforlife/lantern/agents-md"><img src="https://agentmods.dev/badge/instructions/wildchildforlife/lantern/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 | $0.01809 | $0.01809 |
| Opus 5 | $0.00905 | $0.00905 |
| Sonnet 5 | $0.00362 | $0.00362 |
| Haiku 4.5 | $0.00181 | $0.00181 |
Grade A, and why
lantern AGENTS.md scanned grade A 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 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.
Runs shell commandslowCapability
Expected in a hook, worth knowing in a rule or an instructions file.
- `Command.string` not `child_process` Copies of this mod
1 near-identical copy found in the catalogue:
- lantern CLAUDE.md — 86% identical, 71 lines differ
How it starts
The opening of the file, as written. The whole thing — 211 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AGENTS.md
Critical Rules (Read First)
Language:
- Code, comments, commits in English
NEVER:
- Use
astype casting anywhere, tests included (explain problem to user instead) - Use raw
fetchor bypass TanStack Query for API calls - Run
pnpm devorpnpm start(dev servers) - Use
node:fs,node:path, etc. directly (use Effect-TS equivalents)
ALWAYS:
- Effect-TS for all backend side effects
- Hono RPC + TanStack Query for all API calls
- TDD: tests first, then implement
- Run
pnpm typecheckandpnpm fixbefore commit
Commit Message Rules
Conventional Commits: type: description
Release Note Awareness:
- Commit messages ship in release notes. Write for users.
Type Selection:
| Type | Release Note | Purpose |
|---|---|---|
feat |
Features | User-facing new feature |
fix |
Bug Fixes | User-impacting bug fix |
chore, ci, build, refactor |
Excluded | Internal changes |
Critical: fix only for user-facing bugs. Internal fixes (linter errors, type errors, build scripts) use chore.
Message Quality Examples:
- Bad:
fix: fix lingui error(internal issue) - Bad:
feat: add button(too vague) - Good:
feat: add dark mode toggle to settings - Good:
fix: session list not updating after deletion - Good:
chore: update lingui compiled messages
Project Overview
Lantern read Claude Code session logs straight from JSONL files (~/.claude/projects/), zero data loss. Web client shipped as CLI tool serving Vite app.
Core Architecture:
- Frontend: Vite + TanStack Router + React 19 + TanStack Query
- Backend: Hono (standalone server) + Effect-TS (all business logic)
- Data: Direct JSONL reads, strict Zod validation
- Real-time: Server-Sent Events (SSE) for live updates
Recommended Coding Process
Project aims for fast feedback plus quality (checks pass = runtime correctness near-guaranteed) via:
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 · 211 lines · 1,809 tokens per session scan A 49642c9d0c02
lantern AGENTS.md is an instructions file published in the GitHub repository WildChildForLife/lantern (25 stars, last pushed yesterday), licensed MIT. It adds 1,809 tokens to every session, about $0.0090 per session on Opus 5. A static security scan graded it A with 1 finding (runs shell commands). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other instructions, from other repositories
claude-session-dashboard CLAUDE.md
Claude Code instructions for dlupiak/claude-session-dashboard, covering claude.md, mandatory: agent delegation, delegation rules (never skip these), what you (the main context) may do directly and what you must never do directly.
claude-code-hub CLAUDE.md
Claude Code instructions for NikiforovAll/claude-code-hub, covering claude.md, project, commands, architecture and git submodules.
ai-workflow AGENTS.md
Instructions for cunhaax/ai-workflow, covering ai workflow template, rules — non-negotiable, project overview, commands and architecture.
vibetags com-example-config-AsyncTestConfig.instructions.md
Instructions for PIsberg/vibetags, covering copilot instructions for asynctestconfig and immutable type.
pupila CLAUDE.md
Instructions for FranRom/pupila, covering claude.md, where things live, stack, run locally and repo layout.
simon42-dashboard-strategy CLAUDE.md
Claude Code instructions for TheRealSimon42/simon42-dashboard-strategy, covering simon42 dashboard strategy, architecture, module overview, data flow and registry — core design.