🦞 Claw-ESP-Expert

The definitive AI Agent Skill for ESP-IDF Developers.
懂硬件、懂网络、懂工程的专业级 ESP32 开发助理。

Claw-ESP-Expert is a diagnostic-first skill for ESP32 + ESP-IDF workflows. It focuses on the highest-value pain points in embedded development: component lookup, pin safety, build failures, partition overflow, and panic decoding.

Install

npx @movecall/claw-esp-expert

Default behavior:

• installs to ~/.openclaw/workspace/skills/claw-esp-expert
• copies a self-contained skill/ bundle
• does not silently edit SOUL.md
• does not silently modify project files or system permissions

MVP highlights

• Component guidance: query the official ESP Component Registry and generate idf_component.yml suggestions
• Hardware-aware safety: audit risky GPIO assignments across multiple ESP32 chip families
• Build diagnostics: classify common ESP-IDF build failures such as missing headers, component issues, partition overflow, and memory overflow
• Runtime debugging: analyze monitor logs, decode panic backtraces with addr2line, and support a minimal flash + monitor loop

Current scope

Primary tools:

• safe_build
• resolve_component
• analyze_partitions
• decode_panic / analyze_monitor

Supporting tools:

• manage_env
• explore_demo
• flash_and_monitor
• execute_project

Supported chip families:

• esp32
• esp32s3
• esp32c3
• esp32c5
• esp32c6
• esp32h2
• esp32p4

Common SKU names are normalized to family rules, for example:

• ESP32-S3-PICO-1-N8R2 -> esp32s3
• ESP32-C6FH4 -> esp32c6
• ESP32-P4NRW32X -> esp32p4

Local development

git clone https://github.com/movecall/claw-esp-expert.git
cd claw-esp-expert
npm install
npm run build
npm test

Repo layout

claw-esp-expert/
├── bin/          # npx installer
├── skill/        # publishable skill bundle
├── src/          # runtime source
├── scripts/      # build helpers
├── tests/        # smoke tests
├── package.json
└── README.md

Notes

• This is a usable MVP, not a full automation platform
• It returns suggestions, diagnostics, and patch-style drafts before attempting deeper automation
• HIL automation and full self-healing workflows are intentionally out of scope for the first release