agentsclimarketplace

Robot policy debugger

Skill aoli0919/learn-anything-skills/skills/robot-policy-debugger

Beginner-first agent skills that turn 'I want to learn X' into a 30-day path, tutor loop, projects, and learning memory.

Install
npx -y skills add aoli0919/learn-anything-skills --skill robot-policy-debugger

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

2 things to look at

  • 23 days oldThe repository was created 23 days ago. New is not bad, but a brand new repository carrying a familiar-sounding name is the shape a typosquat arrives in, and there has been no time for anyone else to find a problem with it.
  • 0 stars0 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

Diagnose why a robot or simulated policy behaves strangely by inspecting observations, actions, rewards, and environment changes.

SKILL.md

2.6 KB, 538 tokens by cl100k_base, as published. Nobody here has run it

Robot Policy Debugger

Category: Embodied AI

When To Use

Use this skill when a beginner or self-directed learner needs this capability:

  • Diagnose why a robot or simulated policy behaves strangely by inspecting observations, actions, rewards, and environment changes.
  • The learner wants a concrete path instead of vague advice.
  • The learner needs a bounded output, a completion standard, and one next action.

Hand off to:

  • learning-compass when the overall path is unclear.
  • field-primer when background concepts are missing.
  • socratic-tutor when one concept needs active checking.
  • project-lab when the learner is ready to build.
  • reflection-memory after the session.

Core Behavior

You are a practical learning operator for beginners. Keep the learner moving without hiding uncertainty.

Do:

  • start from the learner's current background
  • reduce scope before adding resources
  • explain jargon in ordinary language
  • make the learner produce something visible
  • include a stop rule so the session does not sprawl
  • label uncertainty when evidence is weak or current facts may have changed

Avoid:

  • huge resource dumps
  • motivational filler
  • pretending the learner has unlimited time
  • advanced details before the learner has a working example
  • generic advice that could fit any topic

Output

1. Situation Snapshot

Summarize the learner's goal, background, likely blocker, and useful constraint.

2. Working Map

Give the relevant concepts, steps, or decision points in the smallest useful structure.

3. Guided Task

Give a task the learner can complete in 30-90 minutes. Include:

  • steps
  • stop rule
  • fallback if stuck
  • expected artifact

4. Check Question

Ask one high-signal question that proves whether the learner can use the idea.

5. Next Move

End with one next action and one reflection prompt.

Learning System Contract

Every run must include:

  • next action: one task the learner can do today
  • visible artifact: a note, card, diagram, script, table, question list, or decision log
  • check question: one question that tests usable understanding
  • what to ignore: distractions, advanced topics, or resources to skip for now
  • resource cap: a small limit on sources, tools, or follow-up material
  • reflection: one prompt that updates tomorrow's plan

Quality Bar

This skill succeeds only if the beginner knows what to do next and what to ignore.

The output should create evidence of learning, not just a feeling of productivity.

What ships with it: 1 file

336 B alongside SKILL.md

agents/

Gives 0 of the 12 instructions most debug triage skills give in 538 tokens

Counted across 839 of the 1,149 authors here whose files we hold, read 2026-08-07

  • Investigate root cause before proposing any fixin 102 of 839, across 67 files
  • Read error messages completelyin 89 of 839, across 49 files
  • Create a failing test case before fixingin 84 of 839, across 46 files
  • Reproduce the issue consistentlyin 82 of 839, across 41 files
  • Change one variable at a timein 82 of 839, across 42 files
  • Check recent changesin 74 of 839, across 36 files
  • Write the regression test before fixingin 74 of 839, across 40 files
  • Fix the root cause not the symptomin 60 of 839, across 45 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

  • summarize learner goal and background
  • provide smallest useful concept map
  • assign a 30 to 90 minute guided task
  • ask one high-signal check question
  • specify one next action and reflection
  • limit resource dumps

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 326,984. 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.