From oh-my-team
Spawns agent teams for multi-hypothesis parallel debugging: generates competing bug theories, assigns investigators to test evidence for/against, synthesizes findings, and applies verified minimal fixes.
npx claudepluginhub erkandogan/oh-my-teamThis skill uses the workspace's default tool permissions.
```
Debugs complex bugs using competing hypotheses across 6 failure categories, parallel evidence collection with strength ratings, and root cause arbitration.
Enforces scientific-method debugging loop (Observe→Hypothesize→Experiment→Conclude) for non-trivial bugs like crashes, flaky tests, performance regressions, or CI failures.
Provides systematic debugging framework for root cause analysis after 2+ failed fixes, complex failures, intermittent bugs, and circular debugging.
Share bugs, ideas, or general feedback.
TeamCreate(team_name="debug", description="Multi-hypothesis debugging: $ARGUMENTS")
TaskCreate(subject="Hypothesis 1: [theory]", description="Investigate and find evidence for/against")
TaskCreate(subject="Hypothesis 2: [theory]", description="Investigate and find evidence for/against")
TaskCreate(subject="Hypothesis 3: [theory]", description="Investigate and find evidence for/against")
Agent(
prompt="HYPOTHESIS: [theory 1]
Investigate this theory about: $ARGUMENTS
1. What evidence would CONFIRM this?
2. What evidence would DISPROVE this?
3. Search for both. Don't just confirm.
4. Share findings with other teammates via SendMessage.
5. If disproved, say so clearly.",
subagent_type="explorer",
team_name="debug",
name="investigator-1"
)
Agent(
prompt="HYPOTHESIS: [theory 2] ...",
subagent_type="explorer",
team_name="debug",
name="investigator-2"
)
Agent(
prompt="HYPOTHESIS: [theory 3] ...",
subagent_type="explorer",
team_name="debug",
name="investigator-3"
)
After all investigators report, resolve conflicts:
NEVER apply fixes for multiple hypotheses simultaneously. Fix one, verify, then address the next if needed.