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/mavlink/qgroundcontrol/agents-mdgit clone --depth 1 https://github.com/mavlink/qgroundcontrolWhat 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 | $0.01521 | $0.01521 |
| Opus 5 | $0.00760 | $0.00760 |
| Sonnet 5 | $0.00304 | $0.00304 |
| Haiku 4.5 | $0.00152 | $0.00152 |
Grade A, and why
qgroundcontrol 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 — 110 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AGENTS.md
Instructions for AI coding agents (Codex, Claude Code, etc.) working on QGroundControl.
Quick References
- CODING_STYLE.md — Naming, formatting, C++20 features, QML style, logging
- .github/CONTRIBUTING.md — Architecture patterns (Fact System, Multi-Vehicle, FirmwarePlugin)
- tools/README.md — Development scripts and tooling
- test/README.md — Test framework, base classes, CTest labels, MultiSignalSpy, coverage
- .github/ci-overview.md — CI workflow/action/script layout and conventions
- .pre-commit-config.yaml — All enforced linters (clang-format, clang-tidy, ruff, pyright, shellcheck, actionlint, zizmor, qmllint, clazy, vehicle-null-check, check-no-qassert, check-no-qtest-ignore-message)
Required Preflight
Before modifying any repository file:
- Read CODING_STYLE.md completely.
- Read the task-specific references listed in this file.
- Do not begin editing until those reads are complete.
- Before the first edit, state in a progress update which instruction files were read.
Golden Rules (enforced — violations fail CI)
These are the non-negotiables. The first four are QGC's core architecture patterns; the rest are enforced by pre-commit hooks, so ignoring them wastes a build cycle. Full list with code examples: .github/CONTRIBUTING.md#architecture-patterns and CODING_STYLE.md#common-pitfalls.
- Fact System — ALL vehicle parameters flow through Facts; never create custom parameter storage.
- Multi-Vehicle — ALWAYS null-check
activeVehicle()/Vehicle*before dereferencing (vehicle-null-check). - Firmware Plugin — use
vehicle->firmwarePlugin()for firmware-specific behavior, notif (px4)branches. - QML Integration — register types with
QML_ELEMENT/QML_SINGLETON/QML_UNCREATABLE; expose state viaQ_PROPERTY. - No
Q_ASSERTin production code — use defensive checks with early returns (check-no-qassert). - No
QTest::ignoreMessagein tests — useexpectLogMessage/ignoreLogMessage(check-no-qtest-ignore-message). - No fixed-delay
QTest::qWait(<n>)— useQTRY_*_WITH_TIMEOUTorQSignalSpy::wait(check-no-fixed-qwait). - Comments — prefer self-documenting code; comment only non-obvious intent, constraints, or workarounds.
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 · 110 lines · 1,521 tokens per session scan A 5626f7ebbfa9
qgroundcontrol AGENTS.md is an instructions file published in the GitHub repository mavlink/qgroundcontrol (4,888 stars, last pushed today), licensed Apache-2.0. It adds 1,521 tokens to every session, about $0.0076 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 2026-08-30.
Other instructions, from other repositories
mavlink-mcp CLAUDE.md
Instructions for hieutachi/mavlink-mcp, covering claude.md, what this is, architecture, commands and environment.
ai-drone-toolkit AGENTS.md
AGENTS.md instructions for robotto-xyz/ai-drone-toolkit, covering agents.md — ai drone toolkit, 1. what this project is, 2. repository layout, setup & common commands and 3. non-negotiable principles.
nim_duilib CLAUDE.md
Instructions for rhett-lee/nim_duilib, covering nimduilib - 跨平台 c++ ui 库, 项目概述, 项目结构, 开发模式（xml + c++） and xml 布局文件.
bagel AGENTS.md
Instructions for Extelligence-ai/bagel, covering bagel for agents, set up bagel for a user, use bagel well and develop on bagel.
bagel CLAUDE.md
Instructions for Extelligence-ai/bagel: See AGENTS.md for setup, usage etiquette, and development workflow. It is the single source of truth for agent instructions in this repo.
ha-airspace CLAUDE.md
Instructions for ifnull/ha-airspace, covering claude.md, project at a glance, tech stack, code conventions and style.