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/jrenaldi79/sidecar/claude-mdgit clone --depth 1 https://github.com/jrenaldi79/sidecarWrote 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/jrenaldi79/sidecar/claude-md)<a href="https://agentmods.dev/instructions/jrenaldi79/sidecar/claude-md"><img src="https://agentmods.dev/badge/instructions/jrenaldi79/sidecar/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 | $0.05773 | $0.05773 |
| Opus 5 | $0.02887 | $0.02887 |
| Sonnet 5 | $0.01155 | $0.01155 |
| Haiku 4.5 | $0.00577 | $0.00577 |
Grade A, and why
sidecar 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 — 408 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 when working with code in this repository.
Project Overview
Claude Sidecar is a multi-model subagent tool that extends Claude Code with the ability to spawn parallel conversations with different LLMs (Gemini, GPT-4, o3, etc.) and fold the results back into the main context.
Core Features
- Fork & Fold Workflow: Spawn specialized models for deep exploration, fold summaries back
- Multi-Model Routing: Use the right model for the job (Gemini's large context, o3's reasoning, GPT-4's coding)
- Clean Context: Isolate deep explorations to sidecars, keep main conversation focused
- Async-Safe Operations: File conflict detection and context drift warnings
- Session Persistence: Resume, continue, or read previous sidecar sessions
Key Value Proposition
- Right model for the job - Route tasks to specialized models
- Keep context clean - Isolate deep explorations
- Work in parallel - Background execution with Ctrl+B
- Safe async - Conflict and drift detection
Essential Commands
Development
npm start # Run sidecar CLI
npm run lint # Run ESLint
Testing
npm test # Unit tests (excludes *.integration.test.js)
npm run test:integration # Integration tests only (real LLM, costs tokens)
npm run test:all # Unit + integration (used by pre-push)
npm test tests/context.test.js # Single file (preferred during dev)
npm test -- --coverage # Coverage report
Enforcement
node scripts/check-secrets.js # Scan staged files for secrets
node scripts/check-file-sizes.js # Check staged files against 300-line limit
node scripts/generate-docs.js # Regenerate auto sections in CLAUDE.md
node scripts/generate-docs.js --check # Verify auto sections are current (CI mode)
node scripts/validate-docs.js # Pre-commit: warn if CLAUDE.md may need update
node scripts/validate-docs.js --full # Full: compare CLAUDE.md against codebase
npm run validate-docs # Alias for --full mode
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 · 408 lines · 5,773 tokens per session scan A e23f42434cf5
sidecar CLAUDE.md is an instructions file published in the GitHub repository jrenaldi79/sidecar (17 stars, last pushed 5mo ago), licensed MIT. It adds 5,773 tokens to every session, about $0.0289 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
LLM_trader AGENTS.md
AGENTS.md instructions for qrak/LLM_trader, covering llm trader — master architecture blueprint, 0. instruction authority, 1. system overview, 2. agent inventory and 3. application lifecycle.
vellium AGENTS.md
Instructions for tg-prplx/vellium, covering agents.md — vellium contributor and coding-agent guide, 1. product and runtime model, 2. start every task safely, 3. repository map and renderer: src/.
earheart AGENTS.md
Instructions for cleanunicorn/earheart, covering agents.md, golden rules, the github flow, step by step, 1. start from an up-to-date main and 2. create a branch.
mosaic-companion AGENTS.md
Instructions for hypercycle-development/mosaic-companion, covering mosaic companion — ai agent instructions, project overview, architecture — must read before modifying, key rules and key architecture docs.
garden-desk AGENTS.md
Instructions for alex-alecu/garden-desk, covering agents.md, current phase, test rule, model limitation rule and minimum implementation rule.
OGAD AGENTS.md
AGENTS.md instructions for off-grid-ai/OGAD, covering off grid ai desktop — agent guide, design — desktop-first, off grid brand, use the screen real estate — desktop density rules, what this app is and stack.