agentsclimarketplace

Low power debug

Skill easyzoom/aix-skills/skills/low-power-debug

Reusable Agent Skills for embedded systems, MCU debugging, firmware workflows, and AI automation. （嵌入式、MCU 调试、固件流程与 AI 自动化的可复用 Agent Skills 集合）。

Install
npx -y skills add easyzoom/aix-skills --skill low-power-debug

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

One thing to look at

  • 22 stars22 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

What its author says it does

Copied from the file, not written here

Use when debugging embedded sleep, standby, deep sleep, wakeup, current consumption, clock gating, lost debug access, or low-power mode failures

SKILL.md

3.3 KB, 671 tokens by cl100k_base, as published. Nobody here has run it

Low Power Debug

Overview

Use this skill to debug low-power behavior by measuring current and wake sources before changing sleep configuration. Low-power bugs often involve board leakage, clocks, wake flags, debug settings, peripheral state, and firmware sequencing.

When To Use

Use this skill when:

  • Current consumption is too high or unexpectedly low.
  • The device fails to enter sleep, cannot wake, wakes immediately, or loses debug access.
  • The task involves sleep, stop, standby, deep sleep, retention RAM, wake pins, RTC, watchdog, clock gating, or PMIC sequencing.

Do not use this skill when the board cannot boot at all; establish basic boot first.

First Questions

Ask for:

  • Chip/board and intended low-power mode.
  • Measured current, expected current, and measurement setup.
  • Wake sources and whether wake flags are available.
  • Debug probe attached or detached during measurement.
  • Peripherals left enabled before sleep.
  • Power rails, external devices, pull-ups, and board revision.

Workflow

  1. Confirm measurement validity. Check meter range, sampling rate, burden voltage, shunt placement, rails included, and whether debugger/USB is attached.

  2. Separate firmware from board leakage. Measure baseline board state, external devices, pull-ups, LEDs, regulators, and sensors.

  3. Confirm intended sleep mode. Verify the actual mode entered, not just the API called.

  4. Capture wake reason. Read wake flags immediately after wake before clearing them.

  5. Disable peripherals methodically. Clocks, UART, ADC, timers, DMA, USB, radio, sensors, and pull configurations can dominate current.

  6. Protect debug access. Warn that deep sleep may disconnect SWD/JTAG. Use timed wake, GPIO heartbeat, or connect-under-reset recovery.

Checks

  • Debugger and semihosting disabled for current measurement.
  • Unused GPIO configured to safe states.
  • External pull-ups/pull-downs not causing leakage.
  • Regulator and sensor quiescent currents accounted for.
  • Wake pins polarity and pull state correct.
  • RTC/LPTIM/watchdog clock source active if used.
  • SRAM/retention domains configured intentionally.
  • Interrupt flags cleared before sleep and read after wake.

Verification

Before claiming progress:

  • State measured current, expected current, mode, and measurement setup.
  • State whether debugger/USB was connected.
  • Report actual wake source or wake flag evidence.
  • List peripherals and rails still enabled.
  • State recovery method if debug access may be lost.

Common Failures

  • Measuring through USB while claiming battery current.
  • Leaving the debugger attached and trusting sleep current.
  • Clearing wake flags before reading them.
  • Ignoring external sensors, pull-ups, LEDs, or regulators.
  • Entering a shallower sleep mode than intended because a peripheral clock is active.

Example

User:

进入低功耗后电流还有 5mA，预期几十微安。

Agent:

  1. Asks for mode, current setup, rails measured, debugger state, wake sources, and board peripherals.
  2. Separates board leakage from MCU sleep current.
  3. Checks enabled clocks, GPIO states, regulators, and wake flags before code changes.

Gives 0 of the 12 instructions most debug triage skills give in 671 tokens

Counted across 839 of the 1,149 authors here whose files we hold, read 2026-08-06

  • investigate root cause before proposing any fixin 102 of 839, across 65 files
  • read error messages completelyin 90 of 839, across 48 files
  • create a failing test case before fixingin 84 of 839, across 44 files
  • reproduce the issue consistentlyin 82 of 839, across 40 files
  • change one variable at a timein 82 of 839, across 42 files
  • check recent changesin 74 of 839, across 35 files
  • write the regression test before fixingin 74 of 839, across 36 files
  • fix the root cause not the symptomin 60 of 839, across 43 files
  • implement a single fix at a timein 59 of 839, across 20 files
  • trace data flow backward to the sourcein 50 of 839, across 20 files
  • remove all debug instrumentationin 49 of 839, across 13 files
  • form a single hypothesisin 48 of 839, across 18 files

Said here and by no other author read

  • ask for measurement setup and current expectations
  • separate firmware behavior from board leakage
  • verify the actual sleep mode entered
  • capture wake reason before clearing flags
  • disable peripherals methodically
  • protect debug access before entering deep sleep

Grouped from the skills themselves: near-identical wordings counted once, and counted by distinct author, so one author publishing three of these counts once. Length counted with cl100k_base; the agent that loads this file may tokenize it differently.

Keep looking

Skills are one crate of 328,083. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.