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/auravixstudio/caelo/claude-mdgit clone --depth 1 https://github.com/AuraVixStudio/caeloWrote 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/auravixstudio/caelo/claude-md)<a href="https://agentmods.dev/instructions/auravixstudio/caelo/claude-md"><img src="https://agentmods.dev/badge/instructions/auravixstudio/caelo/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 5.1 | $0.15046 | $0.15046 |
| Opus 5 | $0.07523 | $0.07523 |
| Sonnet 5 | $0.03009 | $0.03009 |
| Haiku 4.5 | $0.01505 | $0.01505 |
Grade A, and why
caelo 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 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.
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 — 603 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.
What this is
Desktop app for Grok (xAI) in the style of Claude Code / Codex: chat, image/video
generation & editing, and an agentic coding module with local file access. It is a
monorepo rebuild of an older customtkinter (Python) app into Electron (frontend) +
Python FastAPI sidecar (backend). The mature xAI logic (OAuth, SSE streaming, media) is
reused, not rewritten. Full plan & phase status live in docs/plans/REBUILD_PLAN.md
(source of truth). The old customtkinter app has been removed from the repo (kept as an external
backup) — Phase 8 closed.
The single most important structural fact
The shared xAI core lives at the repo root, NOT inside caelo_core/:
config.py, api_manager.py, oauth_manager.py, chats_manager.py, history_manager.py,
and make_icon.py. The caelo_core sidecar imports these as top-level modules (import config,
from api_manager import APIManager, …) — caelo_core/__init__.py prepends the repo root to
sys.path at import time so this works, and caelo_core.spec declares them as hiddenimports
with pathex='.'. (The now-removed archive/app.py also reused them via a sys.path shim — hence
they predate the sidecar — but the binding constraint is the sidecar + PyInstaller + data paths.)
Do not move, rename, or restructure these root modules. Doing so breaks caelo_core imports,
the PyInstaller build, and (via config.py) every data-file path. New backend code belongs in
caelo_core/; only touch the root modules to fix shared xAI logic.
Architecture
Electron main (desktop/src/main/index.ts)
• spawns the sidecar (dev: `python -m caelo_core`; packaged: resources/caelo-core/caelo-core.exe)
• generates a session token → CAELO_CORE_TOKEN env; reads handshake line from sidecar stdout
• /health monitor every 10s, auto-restart on crash (≤5 tries); kills sidecar on quit
│ preload (contextBridge) exposes window.caelo → Renderer (React 19 + TS)
▼ HTTP REST + WebSocket — 127.0.0.1 only, bearer/query token
Python sidecar "caelo-core" (FastAPI/uvicorn, caelo_core/)
• server.py mounts routers; state.py Backend wraps reused legacy managers + keys/settings
• agent/ = coding-agent engine (workspace sandbox, permission gate, tools, llm, session loop)
│ Bearer token (OAuth access token → API key → XAI_API_KEY) — only to api.x.ai
▼ xAI / Grok API
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 · 603 lines · 15,046 tokens per session scan A 7f55cd631b38
caelo CLAUDE.md is an instructions file published in the GitHub repository AuraVixStudio/caelo (23 stars, last pushed 23d ago), licensed Apache-2.0. It adds 15,046 tokens to every session, about $0.0752 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
open-codesign CLAUDE.md
Instructions for OpenCoworkAI/open-codesign, covering claude.md — open codesign, what this project is, hard constraints (do not violate), stack & conventions and frontend stack (locked).
open-codesign AGENTS.md
Instructions for OpenCoworkAI/open-codesign, covering agents.md - open codesign, what this project is, hard constraints, ai visibility for web work and current architecture direction.
lfcode AGENTS.md
AGENTS.md instructions for lfyxhappy/lfcode, covering project structure, issue records, build and test, branch names and commits and pr titles.
opencove AGENTS.md
AGENTS.md instructions for DeadWaveWave/opencove, covering system prompt: opencove project ai developer agent, 1. core directives & golden rules, 2. decision framework (small vs large), a. small change (fast feedback / 小步快反馈) and b. large change (deep thinking / 慎重对齐).
wesight AGENTS.md
Instructions for freestylefly/wesight, covering claude.md, build and development commands, build production bundle (typescript + vite), lint with eslint and run memory extractor tests (node.js built-in test runner).
limboo CLAUDE.md
Claude Code instructions for limboo-ai/limboo, covering claude.md, 1. what is limboo?, 2. tech stack (current), 3. project structure and the three electron contexts (critical mental model).