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/arsenyspb/comet-boundary/agents-mdgit clone --depth 1 https://github.com/arsenyspb/comet-boundaryWrote 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/arsenyspb/comet-boundary/agents-md)<a href="https://agentmods.dev/instructions/arsenyspb/comet-boundary/agents-md"><img src="https://agentmods.dev/badge/instructions/arsenyspb/comet-boundary/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 5.1 | $0.00767 | $0.00767 |
| Opus 5 | $0.00383 | $0.00383 |
| Sonnet 5 | $0.00153 | $0.00153 |
| Haiku 4.5 | $0.00077 | $0.00077 |
Grade A, and why
comet-boundary AGENTS.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 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.
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 — 77 lines — stays where its author put it; the contents beside it link to each section on GitHub.
LLM Context Guide (AGENTS.md)
This document serves as the System Prompt Extension for all AI agents (Gemini, Devin, Claude, etc.) operating in the Comet Boundary repository. It prioritizes project-specific technical standards, architectural direction, and operational requirements.
🚨 Critical Guardrails
Before submitting any code or declaring a task complete, you MUST ensure the following:
- Demo Integrity: The environment must be reproducible via
make replay. - Verification: Every change must pass
./scripts/02_verify-setup.sh. - Credential Protection: Never hardcode credentials. Use
.env(backend) orclient/.env.local(frontend). - No "Just-in-Case" Code: Avoid adding features or "cleanup" outside the explicit scope of your task.
🏗️ Architectural Mapping
Build Modes (CRITICAL CONSTRAINT)
Comet Boundary maintains two strictly separate deployment architectures. Never attempt to merge these.
| Mode | Architecture | Purpose |
|---|---|---|
Local Dev (make replay) |
Two containers: Vite frontend (5173) + Go backend (8080) | Preserves Vite HMR for frontend development. |
| Production (GHCR/Helm) | Single container: Go binary embeds React SPA (embed.FS) |
Single artifact distribution via Dockerfile multi-stage build. |
Directory Overview
server/: Go-based SSH/RDP proxy. Uses Chi and the Boundary Go SDK.client/: React + Vite + TypeScript frontend. UI for terminal and RDP canvas.docker/: Infrastructure definitions for Boundary, Postgres, and target systems.scripts/: Critical bootstrap and verification logic.
State & Truth
- Roadmap: The dynamic roadmap is maintained in GitHub Issues.
- Credentials: Infrastructure IDs (Target ID, Auth Method ID) are dynamically generated and injected into
.envfiles byscripts/01_setup-boundary.sh.
🛠️ Common Procedures
1. Resetting the Environment
If the environment becomes inconsistent or unreachable:
make replay
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 · 77 lines · 767 tokens per session scan A 30a3a868808c
comet-boundary AGENTS.md is an instructions file published in the GitHub repository arsenyspb/comet-boundary (5 stars, last pushed 1mo ago), licensed Apache-2.0. It adds 767 tokens to every session, about $0.0038 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 2026-08-31.
Other instructions, from other repositories
vuestrata AGENTS.md
AGENTS.md instructions for boussadjra/vuestrata, covering using vite+, the unified toolchain for the web, built-in commands vs scripts, tool versions, review checklist and vuestrata project rules.
vuestrata CLAUDE.md
Claude Code instructions for boussadjra/vuestrata, covering claude.md, the one thing that trips people up, slash commands, verifying and testing reka-backed components.
clum AGENTS.md
AGENTS.md instructions for tddh/clum, covering clum 项目开发规则, mcp 工具使用规则（最高优先级）, 开发规范, 项目结构 and 常用命令.
BetterDesk copilot-instructions.md
Copilot instructions for UNITRONIX/BetterDesk, covering betterdesk console - instrukcje dla copilota, 📝 konwencja commitów (obowiązkowe), 📊 stan projektu (aktualizacja: 2026-04-10), wersja skryptów all-in-one (v2.4.0) and konsole webowe.
vscode buildNext.instructions.md
Working notes and architecture documentation for the new esbuild-based build system in build/next. Use when making changes to the new build pipeline (transpile/bundle commands, NLS plugin, source-map handling, resource copying, or self-hosting watch tasks).
codex AGENTS.md
AGENTS.md instructions for openai/codex, covering rust/codex-rs, the codex-core crate, code review rules, crate api surface and model visible context.