Oma Debug
Root-cause debugging that enforces a red-then-green regression test on every fix.
Test report
- Verdict
- Works with setup
- Score
- Tested
- Jul 13, 2026
- Environment
- Claude Code 2.x (agent harness)
- Upstream re-checked
- Aug 10, 2026 · 20b9dd8
Fixed a real ZeroDivisionError in a Python helper: the skill's checklist forced verifying the regression test actually failed pre-fix (confirmed) before accepting it, and made me grep for the same unguarded-division pattern elsewhere — rigor a normal ad hoc fix skipped. But most of its resource links (`../_shared/core/*`, `../oma-observability/SKILL.md`, an assumed Serena MCP) reach outside the oma-debug folder into its parent oh-my-agent framework, so copying just this one skill folder leaves several references dangling.
Scored on four weighted criteria — install, triggering, output vs. baseline, docs. How scoring works
- Installs cleanly 4/5
- Triggers reliably 5/5
- Output vs. baseline 8/10
- Docs & honesty 3/5
What Oma Debug does
Diagnoses bugs, crashes, and regressions, applies a minimal root-cause fix, and requires a regression test that fails before the fix and passes after. Triggers on bug/debug/error/crash/traceback/exception/regression language and explicitly defers feature work and general code review to other agents.
How to install Oma Debug
git clone https://github.com/first-fluke/oh-my-agent
cd oh-my-agent
mkdir -p ~/.claude/skills
cp -r .agents/skills/oma-debug ~/.claude/skills/oma-debug
Skills live in ~/.claude/skills/ (global) or .claude/skills/
(per-project). Restart Claude Code after installing.
Commands — how to trigger Oma Debug
-
/oma-debugRoot-cause debugging that enforces a red-then-green regression test on every fix.
It also activates on plain-language prompts like these:
-
This crashes with a weird traceback, find the root cause and fix it -
I have a regression, write a failing test first then fix the bug -
Debug this exception and give me a red then green test for the fix
Frequently asked questions
- Is the Oma Debug skill free?
- Yes. The skill itself is free from first-fluke/oh-my-agent. SkillProof publishes the install command and an independent test verdict at no cost.
- Does Oma Debug work with Claude Code?
- We tested it with Claude Code 2.x (agent harness) on Jul 13, 2026. Verdict: Works with setup. Fixed a real ZeroDivisionError in a Python helper: the skill's checklist forced verifying the regression test actually failed pre-fix (confirmed) before accepting it, and made me grep for the same unguarded-division pattern elsewhere — rigor a normal ad hoc fix skipped. But most of its resource links (`../_shared/core/*`, `../oma-observability/SKILL.md`, an assumed Serena MCP) reach outside the oma-debug folder into its parent oh-my-agent framework, so copying just this one skill folder leaves several references dangling.
- What is the Oma Debug SkillProof Score?
- 8.0/10 — installs cleanly 4/5, triggers reliably 5/5, output vs. baseline 8/10, docs & honesty 3/5.
- How do I install Oma Debug?
- Copy the install command from this page, run it in your terminal, and restart Claude Code. Skills live in ~/.claude/skills/ (global) or .claude/skills/ inside a project.
- Can I use Oma Debug with Cursor, Copilot, Gemini CLI, Codex or other AI tools?
- The SKILL.md format is native to Claude (Claude Code, Desktop, claude.ai). The instructions inside adapt to other assistants: Cursor rules, GitHub Copilot instructions, Windsurf rules, Custom GPTs, AGENTS.md for OpenAI Codex, and GEMINI.md for Google Gemini CLI — our conversion guides cover each, and the free converter on the tools page does the wrapping for you.