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/notnull92/hera-agent-unity/agents-mdgit clone --depth 1 https://github.com/NotNull92/hera-agent-unityWhat 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 | $0.12675 | $0.12675 |
| Opus 5 | $0.06338 | $0.06338 |
| Sonnet 5 | $0.02535 | $0.02535 |
| Haiku 4.5 | $0.01268 | $0.01268 |
Grade C, and why
hera-agent-unity AGENTS.md scanned grade C with 2 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 2d 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.
Downloads and executes remote codehighSupply chain
curl | sh runs whatever the server returns today, which is not necessarily what it returned when this was reviewed.
- `doctor` says binary missing → tell the user to install it (`curl … | sh` or the README link), do not proceed. Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
- `doctor` says binary missing → tell the user to install it (`curl … | sh` or the README link), do not proceed. How it starts
The opening of the file, as written. The whole thing — 809 lines — stays where its author put it; the contents beside it link to each section on GitHub.
hera-agent-unity — Guide for AI Agents
You are an AI coding agent operating in a Unity project that has
hera-agent-unityavailable as a CLI. This document tells you how to use it efficiently. It is meant to be loaded into your project's rules file so every session has it without spending tokens to discover it.Where to put this content.
AGENTS.mdat the project root is the canonical cross-tool agent rules file, standardized by the Agentic AI Foundation (AAIF) under the Linux Foundation (Dec 2025) and adopted by 60,000+ open-source repositories. OpenAI Codex, Claude Code, Cursor, GitHub Copilot, Gemini CLI, and 30+ tools read this file by default.For multi-tool projects, the cleanest pattern is
AGENTS.mdas the single source of truth plus a one-line stub in tool-specific paths. Tool-specific files only matter when a tool requires a different format (Cursor's.mdcYAML frontmatter is the main case).Recommended layout — multi-tool projects:
- Drop the full guide (or its lean subset) into
AGENTS.mdat the project root.- For tools that need their own format, drop a short stub that defers to
AGENTS.md:
CLAUDE.md→> See AGENTS.md.(Claude Code readsAGENTS.mdnatively since late 2025).cursor/rules/hera-agent-unity.mdc→ frontmatter + the same body (Cursor also supports plainAGENTS.mdas a fallback).github/copilot-instructions.md→ repository-wide pointer toAGENTS.md(Copilot uses nearest-file precedence;.github/skills/is for Agent Skills)GEMINI.md+.agents/agents.md+.agents/skills/hera-agent-unity/SKILL.md→ AntiGravity project entry + workspace handoff + on-demand skill.continuerules→ identical bodyPer-tool target paths (2026-current):
Tool Canonical path Notes OpenAI Codex / AGENTS.md-aware tools AGENTS.mdCross-tool standard. Supports layering: ~/.codex/AGENTS.md→ repo root → subtree →AGENTS.override.md.Claude Code AGENTS.md(orCLAUDE.md)Reads AGENTS.mdnatively.CLAUDE.mdstill works for path-scoped rules and imports.Cursor .cursor/rules/*.mdcYAML frontmatter required for activation. .cursorrules(single-file) is deprecated and ignored by Agent mode.GitHub Copilot .github/copilot-instructions.mdNearest-file precedence. Optional: .github/instructions/*.instructions.mdwithapplyTofrontmatter;.github/skills/for Agent Skills.Google AntiGravity GEMINI.md,.agents/agents.md,.agents/skills/*/SKILL.md.agents/is the native workspace extension directory for agents, workflows, and skills.Continue.dev .continuerulesPlain markdown. Other Tool-specific rules file Most accept plain markdown. Two ways to populate the target file:
- Static — copy the matching stub from
examples/rules/(one file per tool, already formatted).- Dynamic — let the CLI generate it from this guide:
# AGENTS.md / CLAUDE.md / Copilot / Continue.dev — compact always-loaded rules hera-agent-unity doctor --agent-rules --compact >> AGENTS.md # Cursor — compact rules with activation frontmatter hera-agent-unity doctor --agent-rules --compact --format cursor > .cursor/rules/hera-agent-unity.mdc # AntiGravity — compact project entry + full on-demand skill hera-agent-unity doctor --agent-rules --compact >> GEMINI.md hera-agent-unity doctor --agent-rules --format antigravity > .agents/skills/hera-agent-unity/SKILL.md
hera-agent-unity is a CLI that drives a running Unity Editor over HTTP. Common uses: execute C# inside the Editor, read console logs, query the active scene, run tests, capture screenshots, batch several commands in one round-trip. Each call is a tool round-trip; response bytes become your input tokens, so reads cost as much as your own writes.
Quick links:
- Full command catalog →
docs/COMMANDS.md - Architecture →
docs/ARCHITECTURE.md - Custom tool authoring →
hera-agent-unity help custom-tools - Setup / install → README
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.
- 2d ago First seen · 809 lines · 12,675 tokens per session scan C a30f9e79c673
hera-agent-unity AGENTS.md is an instructions file published in the GitHub repository NotNull92/hera-agent-unity (28 stars, last pushed 14d ago), licensed Apache-2.0. It adds 12,675 tokens to every session, about $0.0634 per session on Opus 5. A static security scan graded it C with 2 findings (downloads and executes remote code, makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other instructions, from other repositories
unity-cli-skill AGENTS.md
Instructions for niqibiao/unity-cli-skill, covering agents.md, project overview, invoking the cli, lifecycle and setup and command-first principle.
unity-cli-skill CLAUDE.md
Instructions for niqibiao/unity-cli-skill: Project instructions are maintained in and imported from @AGENTS.md.
unity-mcp CLAUDE.md
Claude Code instructions for CoplayDev/unity-mcp, covering claude.md, what this project is, architecture, three layers on the python side and transport modes.
Unity-MCP copilot-instructions.md
Instructions for IvanMurzak/Unity-MCP, covering project guidelines, critical rules, no c# reflection for private access, code review and code style.
mcp-unity AGENTS.md
Instructions for CoderGamester/mcp-unity, covering mcp unity — ai agent guide (mcp package), purpose (what this repo is), how it works (high-level data flow), key defaults & invariants and repo layout (where to change what).
core CLAUDE.md
Instructions for ReactUnity/core, covering claude.md, what this repo is, two package universes, pnpm workspace membership and commands.