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/ghostswap1/ghostswap-agents/agents-mdgit clone --depth 1 https://github.com/ghostswap1/ghostswap-agentsWrote 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/ghostswap1/ghostswap-agents/agents-md)<a href="https://agentmods.dev/instructions/ghostswap1/ghostswap-agents/agents-md"><img src="https://agentmods.dev/badge/instructions/ghostswap1/ghostswap-agents/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 | $0.02321 | $0.02321 |
| Opus 5 | $0.01161 | $0.01161 |
| Sonnet 5 | $0.00464 | $0.00464 |
| Haiku 4.5 | $0.00232 | $0.00232 |
Grade A, and why
ghostswap-agents 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 3d 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 — 149 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AGENTS.md — GhostSwap Partners API agent guide
This file follows the agents.md standard (adopted by OpenAI Codex, Cursor, Gemini CLI, Windsurf, GitHub Copilot, Aider, Continue.dev, Jules, Zed, Devin, JetBrains Junie, and 20+ other tools as of 2026).
Read this when an end-user asks you to add crypto-to-crypto swap functionality to an application, build a swap bot, embed a swap widget, or quote / execute token exchanges programmatically.
What the GhostSwap Partners API does
A server-to-server REST API for non-custodial crypto swaps. End-users send funds to a deposit address you return; the destination wallet receives the swapped funds. GhostSwap handles all upstream liquidity, signing, and clearing — the developer never touches signing keys. Partners earn a 0–4 % markup (set at application time) on every completed swap, paid out in USDT.
Public docs: https://partners.ghostswap.io/docs LLM-friendly index: https://partners.ghostswap.io/llms.txt Full prose walkthrough: https://partners.ghostswap.io/skill.md Machine-readable OpenAPI 3.1 spec: https://partners-api.ghostswap.io/openapi.json
Pick the right surface for your runtime
| Your runtime | What to use | Why |
|---|---|---|
| Claude Code / Claude Desktop / claude.ai | The MCP server in mcp-server/, OR the SKILL.md in skills/ghostswap-partners-api/ |
MCP gives typed tools; SKILL.md gives prose context |
| Cursor / Windsurf / Continue.dev | The MCP server (configure in .cursor/mcp.json / Continue config) |
One-click typed tools |
| ChatGPT Custom GPT, GPT Actions | The OpenAPI spec at openapi/openapi.yaml |
Paste the URL https://partners-api.ghostswap.io/openapi.json into the GPT Actions tab |
| GitHub Copilot coding agent | .github/copilot-instructions.md (symlinked to this file) |
Prose conventions for code generation |
| OpenAI Agents SDK / Vercel AI SDK | The MCP server via MCPClient / experimental_createMCPClient |
MCP is the cross-vendor protocol |
| LangChain / LlamaIndex | The OpenAPI spec via OpenAPIToolkit.from_url(...) or the MCP server via the LangChain MCP adapter |
Either works |
| Aider | This AGENTS.md (auto-loaded if you --read AGENTS.md) plus the OpenAPI spec |
Prose + machine-readable |
| Pasting into any chat (ChatGPT free, Gemini, Perplexity, etc.) | The SKILL.md body (markdown — drop the --- frontmatter or leave it; LLMs skip it) |
Universal markdown |
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.
- 3d ago First seen · 149 lines · 2,321 tokens per session scan A 8401b87eaae1
ghostswap-agents AGENTS.md is an instructions file published in the GitHub repository ghostswap1/ghostswap-agents (3 stars, last pushed 2mo ago), licensed MIT. It adds 2,321 tokens to every session, about $0.0116 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
crypto-claude-skills AGENTS.md
Instructions for Vo1ganin/crypto-claude-skills, covering agents.md, skills, operating rules, read-only by default and credentials.
meeet-solana-state CLAUDE.md
Claude Code instructions for Alvasilev12/meeet-solana-state, covering claude code specific instructions, meeet world — ai agent trust infrastructure on solana and developer portal: meeet.world/developer.
stewie-pixel guidelines
Behavioral guidelines to reduce common LLM coding mistakes. Use when writing, reviewing, or refactoring code to avoid overcomplication, make surgical changes, surface assumptions, and define verifiable success criteria.
behavioral-prediction-mcp CLAUDE.md
Instructions for ChainAware/behavioral-prediction-mcp, covering chainaware behavioral prediction mcp, project overview, mcp tools (14 total), batch pipeline and repository structure.
vanguard-frontier-agentic AGENTS.md
AGENTS.md instructions for VincentChuWaiChow/vanguard-frontier-agentic: The canonical operating guide for this repository is CLAUDE.md. Read it in full before making any change — purpose, layout, working rules, model policy, gates, and definition of done all live there.
vanguard-frontier-agentic GEMINI.md
Gemini CLI instructions for VincentChuWaiChow/vanguard-frontier-agentic: The canonical operating guide for this repository is CLAUDE.md. Read it in full before making any change — purpose, layout, working rules, gates, and definition of done all live there.