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/metaspartan/cybara/claude-mdgit clone --depth 1 https://github.com/metaspartan/cybaraWrote 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/metaspartan/cybara/claude-md)<a href="https://agentmods.dev/instructions/metaspartan/cybara/claude-md"><img src="https://agentmods.dev/badge/instructions/metaspartan/cybara/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 | $0.01479 | $0.01479 |
| Opus 5 | $0.00740 | $0.00740 |
| Sonnet 5 | $0.00296 | $0.00296 |
| Haiku 4.5 | $0.00148 | $0.00148 |
Grade A, and why
cybara CLAUDE.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 6d 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.
### Use Bun.spawnSync instead of child_process How it starts
The opening of the file, as written. The whole thing — 250 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AI Coding Rules for Cybara Agent Platform
This document contains coding standards and best practices for AI assistants working on this codebase.
No Code Comments (STRICT)
Do NOT write comments in code. No //, /* */, #, JSDoc, or docstring comments — not for explaining "why", not for section headers, not for TODOs, not in TypeScript, tests, scripts, Swift, or config. Write self-explanatory code with clear names instead. Never reference competitors or other products in code or comments. This is a hard rule: if a change adds a comment, remove it before finishing. Existing comments may stay; do not add new ones.
Runtime Environment
- Runtime: Bun (not Node.js)
- Package Manager: bun
- TypeScript: Strict mode enabled
Import Rules
✅ DO
// Static imports at top of file
import { homedir } from "os";
import { readFileSync, existsSync, statSync } from "fs";
import { tables } from "./database";
// No .js extension in imports (ESM standard)
import { config } from "../core/config";
❌ DON'T
// No require() - use import instead
const fs = require("fs"); // BAD
// No require() for os.homedir()
process.env.HOME || require("os").homedir(); // BAD
// Prefer:
import { homedir } from "os";
process.env.HOME || homedir(); // GOOD
Dynamic Imports (when acceptable)
Dynamic imports are acceptable ONLY for:
- Circular dependency avoidance - when two modules import each other
- Optional dependencies - like Playwright which may not be installed
- Lazy loading at startup - to speed up initial load time
// Acceptable: Optional dependency
const { chromium } = await import("playwright");
// Acceptable: Circular dependency avoidance
const { listSessions } = await import("../api/chat");
Bun Native APIs
Use Bun.spawnSync instead of child_process
// ❌ DON'T
import { execSync } from "child_process";
const output = execSync("which rg", { encoding: "utf-8" });
// ✅ DO
const result = Bun.spawnSync(["which", "rg"]);
const output = result.stdout.toString();
// For shell commands:
const result = Bun.spawnSync(["sh", "-c", "complex | shell | command"]);
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.
- 6d ago First seen · 250 lines · 1,479 tokens per session scan A 28ad1c3df58d
cybara CLAUDE.md is an instructions file published in the GitHub repository metaspartan/cybara (26 stars, last pushed today), licensed MIT. It adds 1,479 tokens to every session, about $0.0074 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
Ultimate-Agent-Directory AGENTS.md
AGENTS.md instructions for moshehbenavraham/Ultimate-Agent-Directory, covering agents.md, important rule, repository overview, essential build commands and data architecture.
zerostack AGENTS.md
AGENTS.md instructions for gi-dellav/zerostack, a project described as: Lightweight coding agent written in Rust, optimized for memory footprint and performance.
kodama AGENTS.md
Instructions for amergrgic/kodama, covering agents, architecture, agent roster, conventions and validation.
agora AGENTS.md
AGENTS.md instructions for CopperEagle/agora, covering agents.md — the agora, technology decisions (settled), architecture constraints, testing requirements and unit tests (fast, no mcp transport).
truss copilot-instructions.md
Copilot instructions for KornLabs/truss: This project uses the open agents.md standard. Read AGENTS.md — that file is the complete boot instruction for this workspace.
Ultimate-Agent-Directory CLAUDE.md
Claude Code instructions for moshehbenavraham/Ultimate-Agent-Directory, a project described as: 🤖 The most comprehensive directory of AI agent frameworks, platforms, tools, and resources - hundreds of curated entries covering open-source, no-code, enterprise, and autonomous solutions. NEW Boilerplates!