| name | obv - Obvious Check |
| description | Add an obvious-things-first step to any analysis. Catches what everyone assumes someone else checked. |
| output | {"format":"prose"} |
Obvious Check
Input: $ARGUMENTS
Purpose
The most common failure mode in analysis is missing the obvious. Not because it's hard to see — because everyone assumes someone already checked it. This skill adds the "did anyone actually look?" step.
Can be applied standalone or injected into any other skill's output.
Step 1: STATE THE SITUATION
What is being analyzed, decided, or planned?
SUBJECT: [one sentence]
Step 2: THE OBVIOUS QUESTIONS
Answer each. If you can't answer confidently, that's a finding.
A. Obvious Facts
- What is the single most basic fact about this situation? Is it confirmed?
- What would a newcomer ask first? Has that been answered?
- What does everyone involved "know" but nobody has verified?
- Is the problem actually what we think it is? (Have we checked?)
B. Obvious Actions
- What is the simplest thing that could work? Has it been tried?
- What would cost almost nothing to try? Has it been tried?
- Is there a standard solution that already exists? Have we looked?
- Has someone already solved this exact problem? Have we checked?
C. Obvious Risks
- What is the most likely thing to go wrong? Is there a plan for it?
- What happens if we do nothing? Is that actually bad?
- What is the most obvious way this fails? Are we protected?
- Is there a deadline or constraint we're ignoring?
D. Obvious Oversights
- Who haven't we talked to that we obviously should?
- What information do we not have that we obviously need?
- What assumption are we making that we haven't stated?
- What are we optimizing for? Is that the right thing?
Step 3: OBVIOUS VERDICT
OBVIOUS CHECK RESULTS:
Confirmed obvious:
- [things that were checked and are fine]
Unchecked obvious:
- [things nobody actually verified]
Obvious actions not taken:
- [simple things that haven't been tried]
Obvious risks unaddressed:
- [likely failure modes with no plan]
STOP SIGNAL: [YES if any unchecked obvious item could change the conclusion / NO if basics are covered]
If STOP SIGNAL is YES: Address the unchecked items before continuing with more sophisticated analysis.
Step 4: INJECTION MODE
When used as a step within another skill, output only:
OBVIOUS CHECK:
- [x] [checked item — confirmed]
- [ ] [unchecked item — NEEDS ATTENTION]
- [ ] [unchecked item — NEEDS ATTENTION]
Proceed: [YES / NO — address unchecked items first]
Integration
Use with:
- Any skill — inject as a first step before deeper analysis
/ogo -> Specifically check for obvious good outcomes/obo -> Specifically check for obvious bad outcomes/oba -> Full obvious-everything scan/rca -> When the obvious cause hasn't been ruled out
