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 commands/jzl-maker/ea-skill/sigit clone --depth 1 https://github.com/jzl-maker/EA-SKILLWhat 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 | $0.00000 | $0.00908 |
| Opus 5 | $0.00000 | $0.00454 |
| Sonnet 5 | $0.00000 | $0.00182 |
| Haiku 4.5 | $0.00000 | $0.00091 |
Grade A, and why
si 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.
What it actually says
命令: /ea si (存量项目接入)
功能
分析现有嵌入式工程，重建项目状态：建立 <STATE_DIR>/（.ea/ 或沿用老 .em/）+ 完整 context.md + 基线快照，让老项目进入可被 AI 增量开发/调试的状态。
触发
/ea si [路径]
不带路径时使用当前目录。
执行流程
- 【进入目录】 定位项目根
- 【状态目录】
get_state_dir()：已有.ea//.em/→ 沿用并合并；否则新建.ea/ - 【代码审计 + 上下文构建】 走
workflows/context-build.md：- 芯片识别：
system_*.c/startup_*.s→ chips.json 学习（复用workflows/chip-learning.md正则） - 外设识别：搜索
gpio_init/usart_init/i2c_init等关键词 - 构建入口：扫描
*.uvprojx/*.uvproj/*.ioc/sdkconfig/platformio.ini - 模块地图：main / 驱动层 / 应用层 / 中断处理 / printf 重定向 / LED·按键调试接口
- 既有约定：编码（GB2312 检测）、命名规范、架构约束
- 保护区清单：
startup_*.s、向量表、*.sct/*.ld/*.icf、system_*.c - 全部写入
<STATE_DIR>/context.md
- 芯片识别：
- 【基线快照】
python ~/.claude/skills/ea-skill/tools/shared/project_guard.py --snapshot- 保护区 + 关键源文件 SHA-256 → context.md 基线区
- 老工程保护从此生效：之后任何改动都能 diff / 审计
- 【git 检查】
git log -1读最后一次提交信息写入 state.md - 【生成规格】
<STATE_DIR>/project-spec.md步骤表（从代码推断已完成步骤）+state.md - 【输出审计报告】
输出格式
🔍 代码审计完成
📟 芯片: GD32F407VET6 (GigaDevice) 外设: GPIO/UART/I2C/SPI/CAN/Timer
🔧 工程: Keil (.uvprojx) target=GD32F407 linker=xxx.sct
🗺 模块: main → task/Display/Drv_UART... printf→串口3 调试: LED_PC13/按键KEY0
🛡 保护区: startup_gd32f407.s / system_gd32f4xx.c / xx.sct 已加基线
✅ 已完成: S1 板级驱动 S2 显示 ...
🔄 当前状态: 主循环 + 看门狗 250ms；上次提交: <git log 一行>
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
下一步: /ea rec 或 /ea new <功能描述>
设计原则
- ✅ 老工程零破坏：不改任何源文件，只建状态/上下文
- ✅ 增量开发前提：context + 基线让后续所有改动可 diff、保护区可审计
- ✅ 既有
.em/项目兼容：沿用不迁移
相关文件
workflows/context-build.md— 上下文感知构建（本命令核心）workflows/chip-learning.md— 芯片识别/学习tools/shared/project_guard.py— 基线快照templates/context.md— context 模板
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 · 56 lines · 0 tokens per session scan A 00c810334bf2
si is a command published in the GitHub repository jzl-maker/EA-SKILL (6 stars, last pushed 5d ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 908 tokens. 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 commands, from other repositories
fix_board
Automatically diagnose and fix PlatformIO board upload/monitor issues.
expert_rmt5
ESP-IDF v5.x RMT5 API expert for LED protocols, encoders, and multi-channel control.
esp-teach
One-time project setup -- discover hardware, find datasheets, persist context to CLAUDE.md.
esp-debug
Help debug ESP32 crashes, hangs, and peripheral issues.
comm-bus
You are a bus-driver-engineer agent using deep expertise across I2C, SPI, UART, CAN, and USB. Help the user design a correct driver, debug a misbehaving driver, or migrate a driver between MCU families.
iot
You are an iot-protocol-engineer agent with deep expertise across MQTT, CoAP, LwM2M, BLE GATT, LoRaWAN, Thread, and Zigbee. Help the user choose the right protocol for their constraints and design the protocol layer correctly.