Instructions file
Repository maintenance rules for auto-embedded, a TypeScript and Python framework that installs reusable agent workflows and tools across coding platforms.
Instructions file
Repository maintenance rules for auto-embedded, a TypeScript and Python framework that installs reusable agent workflows and tools across coding platforms.
Skill Claude CodeCodex
A project-level framework for developing embedded firmware with AI coding tools. It adds staged planning, shared project specifications, role-based work, architecture and hardware checks, reusable tool skills, and records of lessons learned.
Agent
A builder role for embedded-software projects that follows a written implementation plan. It edits the specified files while respecting layer boundaries, fixed hardware assignments, naming rules, interrupt-service routines, and shared-data safeguards.
Agent
A read-only research role for embedded-device projects, guided by project specifications. It identifies the chip platform, firmware libraries, existing drivers, and official examples.
Agent
A read-only software review role for checking whether an implementation meets written requirements and hardware rules.
Agent
Use when implementing application-layer logic for embedded competition: state machines, CLI command parsers, control laws, signal codecs, file system service modules. Consumes drivers from embedded-drv and gain/coefficient .h files from embedded-matlab. Never touches hardware registers directly.
Agent
Use when starting any embedded competition project. Reads contest problem, routes to task type (MAIN+TAGS), dispatches 4-6 specialist subagents, writes hardware/interface contracts, manages decision gates, and integrates final deliverables. Vision tasks are handed off to the separate auto-vision skill. Always the…
Agent
Use when implementing low-level peripheral drivers (GPIO/UART/SPI/I2C/ADC/RTC/DMA/PWM) for embedded competition projects. Strictly follows hardware contract from embedded-arch, writes Driver/BSP/HAL Port layers, never touches application logic.
Agent
Use when designing algorithms for embedded competition: LQR/PID controllers, IIR/FIR filters, FFT analysis, Kalman observers, DDS LUTs, system identification. Runs MATLAB via mcpmatlab tools, exports .h via . Skipped for SYSTEM-type problems with no algorithm TAGS.
Agent
Use when verifying embedded competition project at CP-3: static analysis, MIL/SIL/PIL three-tier validation, 5-tuple scoring checklist verification, and one-click firmware pipeline. Independent verifier — never implements code, only audits.
Agent
Use when preparing competition report, presentation slides, and defense rehearsal. Reads all other subagents' artifacts and produces LaTeX/Word report, PowerPoint, and 10-why defense Q&A with evidence chains. Mandatory CP-4/CP-5 subagent.
Command
A command for resuming an unfinished auto-embedded development task. It reloads the project context and identifies the current stage, recent changes, hardware-resource state, research findings, and next action.
Command
A project wrap-up command for the auto-embedded framework, which supports structured embedded-software development. It runs automated checks, guides a three-part review, records reusable lessons and session notes, and archives the finished task.
Command
A command that appends a short session summary to workspace/journal.md, a file used to keep notes between coding-agent sessions. New sessions can use recent entries to understand what happened previously.
Command
A command for creating a task and starting its RESEARCH stage in a defined five-step workflow.
Command
A command that prints the current state of an auto-embedded project, including its active task, development phase, specification layer, hardware lock, and developer.