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/amxv/agentbox/agents-mdgit clone --depth 1 https://github.com/amxv/agentboxWhat 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 | $0.01111 | $0.01111 |
| Opus 5 | $0.00556 | $0.00556 |
| Sonnet 5 | $0.00222 | $0.00222 |
| Haiku 4.5 | $0.00111 | $0.00111 |
Grade A, and why
agentbox 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 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.
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 — 131 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Agentbox Agent Guide
Agentbox is a Go service and CLI with independent web and Raycast clients. Keep component boundaries obvious and use the smallest relevant verification loop while editing, then run the complete suite before shipping cross-cutting work.
Repository map
cmd/api/ Go backend executable and backend deployment config
cmd/agentbox/ native CLI executable
cmd/migrate/ explicit database migration command
internal/agentbox/ canonical product implementation shared by backend and CLI
migrations/ embedded ordered PostgreSQL migrations
apps/dashboard/ Next.js user/admin dashboard and current public web surfaces
apps/raycast/ independent Raycast extension
packaging/cli/ @amxv/agentbox npm distribution wrapper
tests/integration/ cross-component dashboard/backend contracts
The MCP server is part of the Go backend at /api/mcp; it is not a separate service. The dashboard proxies its /api/* routes to the Go backend. Raycast and the CLI are API clients of the same backend.
Within the Go implementation, follow the domain filenames instead of starting from a giant catch-all file:
internal/agentbox/db/repository_*.gocontains PostgreSQL behavior by domain (threads,uploads,credentials,accounts,teams); the matchingmemory_*.gofiles mirror those behaviors for the in-memory repository.repository_helpers.goowns shared row scanners/value helpers, whilerepository.goowns the shared repository core.internal/agentbox/service/{threads,uploads,credentials,accounts,sessions}.gocontains product behavior;service.goowns the repository contract, shared types, and cross-domain helpers.internal/agentbox/httpapi/server_{threads,credentials,auth,owner}.gocontains HTTP handlers;server.goowns routing, middleware, parsing/error helpers, and response shaping.- Large Go tests are split by the same durable behavior areas. Prefer adding a test beside the behavior it exercises rather than growing a generic test file.
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 · 131 lines · 1,111 tokens per session scan A bd56ed3ce494
agentbox AGENTS.md is an instructions file published in the GitHub repository amxv/agentbox (1 stars, last pushed 5d ago), licensed Apache-2.0. It adds 1,111 tokens to every session, about $0.0056 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
azure-search-openai-demo bicep.instructions.md
Infrastructure as Code with Bicep.
microfeed AGENTS.md
AGENTS.md instructions for microfeed/microfeed, covering repository guidance, development workflow, instance management and cloudflare deployment, source architecture and theme repositories.
einbiogpt copilot-instructions.md
Instructions for ehsanghaffar/einbiogpt, covering biogpt copilot instructions, project architecture, rate limiting & error handling, key patterns & conventions and environment variables.
einbiogpt AGENTS.md
Instructions for ehsanghaffar/einbiogpt, covering agents.md - coding guidelines for ai agents, project overview, build/development commands, development server and production build.
business-card-mcp AGENTS.md
AGENTS.md instructions for ai-cooperation/business-card-mcp, covering agents.md — business card mcp 部署引導（跨 cli）, 前置, 步驟, 1. 說明與同意（確認 1） and 2. 組裝.
aws-mcp-gateway AGENTS.md
AGENTS.md instructions for rafaself/aws-mcp-gateway, covering agents.md, repository purpose, workflow, development commands and code organization.