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/gcwing/sparo-os/agents-mdgit clone --depth 1 https://github.com/GCWing/Sparo-OSWrote 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/gcwing/sparo-os/agents-md)<a href="https://agentmods.dev/instructions/gcwing/sparo-os/agents-md"><img src="https://agentmods.dev/badge/instructions/gcwing/sparo-os/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 | $0.02819 | $0.02819 |
| Opus 5 | $0.01409 | $0.01409 |
| Sonnet 5 | $0.00564 | $0.00564 |
| Haiku 4.5 | $0.00282 | $0.00282 |
Grade B, and why
Sparo-OS AGENTS.md scanned grade B with 1 finding 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.
Sends data to an external URLmediumData exfiltration
A POST to an outside endpoint may be telemetry or may be exfiltration; either way the mod talks to somewhere, and you should know where.
fetch('http://127.0.0.1:7242/ingest/session-id', { method: 'POST', How it starts
The opening of the file, as written. The whole thing — 238 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Agents.md
中文 | English
Project Overview
Sparo OS is an Agentic OS for building and running intelligent apps. The first-class product surfaces are the Tauri desktop app and the CLI, both backed by shared Rust core services. The React/TypeScript Web UI is the desktop app surface.
Focus routine development on:
src/apps/desktop- Tauri 2 desktop shell, commands, capabilities, and desktop-only integration.src/apps/cli- CLI command surface, terminal/TUI rendering, and CLI-only integration.src/web-ui- React 18 + TypeScript UI used by the desktop app.src/crates/core- platform-agnostic business logic, agent runtime, services, storage, paths, and tools.src/crates/events- platform-agnostic event contracts.src/crates/transport- adapters between core/events and app surfaces.
Do not describe or design features around a general-purpose app server target unless the user explicitly asks for it. The supported product paths are desktop + Web UI and CLI; Remote Connect relay is a separate infrastructure crate, not a Sparo app server.
Current Architecture
src/crates/core/src/agentic- agents, prompts, sessions, dialog turns, model rounds, and tool execution.src/crates/core/src/command- host-agnostic command services shared by desktop and CLI adapters.src/crates/core/src/runtime- shared process and agentic runtime builders used by desktop and CLI.src/crates/core/src/service- workspace, config, filesystem, terminal, git, and related services.src/crates/core/src/infrastructure- AI adapters, app paths, logging, storage, debug ingest, and events.src/web-ui/src/app- application shell and desktop panels.src/web-ui/src/flow_chat- chat UI, tool cards, streaming/tool event presentation.src/web-ui/src/tools- feature tools such as editor, terminal, git, mermaid, and design canvas.src/web-ui/src/infrastructure- theme, i18n, config, API adapters, and state wiring.src/web-ui/src/design-system- reusable UI APIs, visual contracts, preview coverage, and AI-facing UI rules.src/web-ui/src/shared- shared frontend services, markdown rendering, utilities, and types.src/web-ui/src/locales- translations.
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 · 238 lines · 2,819 tokens per session scan B 439ef575767a
Sparo-OS AGENTS.md is an instructions file published in the GitHub repository GCWing/Sparo-OS (20 stars, last pushed 24d ago), licensed MIT. It adds 2,819 tokens to every session, about $0.0141 per session on Opus 5. A static security scan graded it B with 1 finding (sends data to an external url). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other instructions, from other repositories
agentic-os AGENTS.md
Instructions for modimihir07/agentic-os, covering agentic os — complete project context for ai agents, role definition, project identity, architecture and 3-agent engine.
empirica copilot-instructions.md
Copilot instructions for EmpiricaAI/empirica, covering empirica system prompt - copilot v1.6.6, identity, vocabulary, two axes: workflow vs thinking and workflow phases (mandatory).
rome CLAUDE.md
Claude Code instructions for rome-os/rome, covering rome, playbook and traps.
agent-plugins-skills copilot-instructions.md
Copilot instructions for richfrem/agent-plugins-skills, covering copilot instructions for agent-plugins-skills, 1. think before coding, 2. simplicity first, 3. surgical changes and 4. goal-driven execution.
rustyhand AGENTS.md
AGENTS.md instructions for ginkida/rustyhand, covering repository guidelines, project structure & module organization, build, test, and development commands, coding style & naming conventions and testing guidelines.
revaya-aios-workspace-template CLAUDE.md
Instructions for revaya-ai/revaya-aios-workspace-template, covering claude.md, what this is, the claude-user relationship, workspace structure and commands.