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/naorsabag/openhop/agents-mdgit clone --depth 1 https://github.com/naorsabag/openhopWrote 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/naorsabag/openhop/agents-md)<a href="https://agentmods.dev/instructions/naorsabag/openhop/agents-md"><img src="https://agentmods.dev/badge/instructions/naorsabag/openhop/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 | $0.01450 | $0.01450 |
| Opus 5 | $0.00725 | $0.00725 |
| Sonnet 5 | $0.00290 | $0.00290 |
| Haiku 4.5 | $0.00145 | $0.00145 |
Grade C, and why
openhop 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 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.
Recursive force deletehighDestructive command
rm -rf with a variable or a broad path is one typo away from removing the wrong tree.
- `env: can't execute 'tsx'` from `openhop` → you're on an old checkout; `git pull && rm -rf node_modules && npm install`. Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
curl -s http://localhost:8787/health # expect {"status":"ok"} How it starts
The opening of the file, as written. The whole thing — 103 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AGENTS.md
Instructions for AI agents (Claude Code, Cursor, Cline, Codex, OpenAI Swe-agent, etc.) working in this repo.
Humans: this is a companion to README.md — that file has the install walkthrough, screenshots, and feature list. Read it first if you're new here.
What this is
OpenHop is a YAML-driven data-flow visualizer. Authors describe a flow in YAML, push it with the openhop CLI, and the UI animates data pixels traveling between nodes on a pixel-art canvas.
packages/server— Fastify API (:8787), Zod-validated flow store, Swagger at/docs.packages/web— React + React Flow + ELK layout. Served by Vite at:8788.packages/cli—openhopcommand (esbuild-bundled, no runtime TS).packages/shared— Zod schemas + types, imported by all three.skills/openhop/SKILL.md— the canonical authoring guide for any SKILL-compatible agent. If you're being asked to author a flow, load that file first.examples/— reference flows (order-flow.yamlis the canonical example;type-variants.yamlexercises every node type).
Run locally
npm install # installs deps + bundles packages/cli/dist/index.js
npm run dev # API on :8787, UI on :8788
Smoke test before you do anything else:
curl -s http://localhost:8787/health # expect {"status":"ok"}
The dev server hot-reloads on edits under packages/web/src. Server edits restart via tsx watch.
CLI
(cd packages/cli && npm link) # one-time: makes `openhop` global
openhop push <file.yaml> # push a flow
openhop list # list all flows
openhop patch <id> <file.yaml> # update a flow
openhop remove <id> # delete
All commands accept -s <url> to target a non-default server.
Typecheck / build
# per-package
npx tsc --noEmit -p packages/shared
npx tsc --noEmit -p packages/server
npx tsc --noEmit -p packages/web
npx tsc --noEmit -p packages/cli
# CLI bundle (runs automatically via `prepare` on install)
npm run build -w @openhop/cli
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 · 103 lines · 1,450 tokens per session scan C e6f964c89e08
openhop AGENTS.md is an instructions file published in the GitHub repository naorsabag/openhop (42 stars, last pushed yesterday), licensed MIT. It adds 1,450 tokens to every session, about $0.0072 per session on Opus 5. A static security scan graded it C with 2 findings (recursive force delete, 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
oh-my-mermaid CLAUDE.md
Instructions for oh-my-mermaid/oh-my-mermaid, covering project rules, readme localization, versioning, commits and terminology.
diagram-creator-skill AGENTS.md
AGENTS.md instructions for ferdinandobons/diagram-creator-skill, covering agents.md, repository overview, repository structure, skill structure and build / lint / test.
claude-code-architecture CLAUDE.md
Instructions for KhaiTrang1995/claude-code-architecture, covering claude.md, tổng quan dự án · project overview, cấu trúc dự án · project structure, lệnh thường dùng · common commands and scaffold dự án mới · scaffold a new project.
diagram-creator-skill CLAUDE.md
Claude Code instructions for ferdinandobons/diagram-creator-skill, covering diagram creator skill, about this project, key directories, commands and standards.
Paperlab AGENTS.md
AGENTS.md instructions for NourMtir0722/Paperlab, covering paperlab — for coding agents, integrating paperlab into a project, stage mode — paper as architecture, content types and lighting is data, not an enum.
markdy-com copilot-instructions.md
Copilot instructions for HoangYell/markdy-com, a project described as: 🎬 Open-source animated architecture diagrams as code. Diagram-native DSL, 17 layout engines, Web Animations API, and MCP server for AI agents.