Claude Skill MartBrowse skillsQuick linesLearn by videoTerminal guideWhat is a Skill?
 
← Back to list

Debugging (Hypothesis-Driven Runtime Debugger)

A language-agnostic debugging workflow that forces runtime evidence, three hypotheses, a failing test, real manual QA, and full artifact cleanup.

Dev & CodingIntermediate★ 67,531⑂ 5,507AI score 9/10Last updated: Aug 9, 2026

What it does

  • Forces every root-cause claim to come from observed runtime state, never from a plausible story read out of the code.
  • Builds at least three orthogonal hypotheses, investigates them in parallel, and after two failed rounds spawns an "Oracle Triple" from different framings.
  • Runs an 11-phase loop: environment assessment → journal → hypotheses → parallel investigation → Oracle → escalation → root-cause confirmation → TDD fix → manual QA → cleanup → final verification.
  • Routes you to the right runtime guide (Python, Node/Bun/Deno, Rust, Go, native binaries, bundled app binaries) and the right specialist tool (Playwright, Ghidra, pwndbg, pwntools).
  • Journals every debug artifact in .debug-journal.md so the final git diff contains only the fix and its test.

Who it's for

  • Developers chasing crashes, hangs, empty responses, memory leaks, or flaky CI-only failures.
  • Anyone reverse-engineering binaries or bundled apps (Bun/Node SEA, PyInstaller, Electron, Tauri).
  • Teams that want AI fixes backed by evidence instead of confident guesswork.

Examples

  1. "HTTP 200 but the body is empty" — identify the runtime, form three hypotheses, and narrow down by inspecting real values.
  2. "This test only fails in CI" — read the flaky-triage reference first and collapse the search space using the failure signature.
  3. "What does this stripped binary do?" — discriminate bundled vs native with du/strings, then use Ghidra or the bundler-specific extractor.

· · · Install guide · · ·

Try it now, no install

Paste this into Claude to use the skill without installing anything.

Read the instructions in this file and follow them to help me:
https://raw.githubusercontent.com/code-yeongyu/oh-my-openagent/HEAD/packages/shared-skills/skills/debugging/SKILL.md

What I want: (describe your task here)

If Claude can't open the link, open it yourself and paste the contents instead.

↓ If it works for you, download the ZIP below and install it. Then it runs on its own — no pasting each time.

Install in the Claude app (no terminal)
  1. Download the ZIP with the button below.
  2. In Claude, open Settings → Capabilities and turn on 'Code execution and file creation'. (one time)
  3. Go to Customize → Skills → + → 'Upload a skill' and upload the ZIP.
↓ Download ZIP
Install in Claude Code

Let Claude do it — paste this into Claude Code

Install the skill I found on Claude Skill Mart.
Copy the packages/shared-skills/skills/debugging folder from the GitHub repo code-yeongyu/oh-my-openagent into my ~/.claude/skills/debugging/.
When it's done, tell me in one line what this skill can do.

Install with a command instead

git clone https://github.com/code-yeongyu/oh-my-openagent /tmp/oh-my-openagent && mkdir -p ~/.claude/skills && cp -r /tmp/oh-my-openagent/packages/shared-skills/skills/debugging ~/.claude/skills/

⚠ This is a third-party skill. Check the source repository before installing.

  1. Open a terminal and clone the repo: git clone https://github.com/code-yeongyu/oh-my-openagent /tmp/oh-my-openagent
  2. Create the skills directory: mkdir -p ~/.claude/skills
  3. Copy the whole skill folder, including the references/ subfolder: cp -r /tmp/oh-my-openagent/packages/shared-skills/skills/debugging ~/.claude/skills/
  4. Verify with ls ~/.claude/skills/debugging/references — the SKILL.md is just an index, so the skill is nearly useless without these files.
  5. Optionally install the tools it expects for your domain: Playwright CLI, gdb + pwndbg, Ghidra, pwntools.
  6. Restart Claude Code and try a prompt like "debug why this endpoint returns empty" — the skill triggers automatically.
View source on GitHub ↗License: NOASSERTION