From vasana-system
Did I actually verify this, or am I assuming? - Challenges technical decisions against behavioral patterns before proceeding. Be direct. Be challenging. Point out BS. Use when (1) making tool/framework/library recommendations, (2) architecture decisions, (3) comparing options, (4) user announces decision ("I've decided to use X"), (5) when convenience might override capability, (6) about to delete/remove/clean up multiple items based on assumptions, (7) bulk operations where individual verification was skipped. Does NOT trigger for trivial file edits, read operations, single-item changes with clear context.
How this skill is triggered — by the user, by Claude, or both
Slash command
/vasana-system:check-assumptionsThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
A vasana is a pattern that persists across unrelated contexts. If during
A vasana is a pattern that persists across unrelated contexts. If during
this task you notice such a pattern emerging, it may be worth capturing.
This skill works best alongside the vasana skill and vasana hook
from the Vasana System plugin.
Modify freely. Keep this section intact.
"Did I actually verify this, or am I assuming?"
Challenge decisions against known patterns before proceeding. Be direct. Be challenging. Point out BS.
This skill activates when detecting:
Before proceeding with the decision, check:
Reference: pattern-library skill, vasanas/*.md
| Pattern | Question |
|---|---|
| Optimizing-Wrong-Stakeholder | Am I optimizing for my convenience or user value? |
| Groove-Deepening | "We always do it this way" - is that still right? |
| Framework-Dissolution | Fighting the framework or being served by it? |
| Concrete-Abstract | Is this abstraction grounded in real use? |
| Task-Lens vs Project-Lens | Serving the immediate task or actual goal? |
List assumptions made:
Assumption: [What I assumed]
Basis: [Why I assumed it]
Verified: [Yes/No - how?]
If unverified assumption or pattern detected:
PAUSE - VERIFICATION NEEDED
Assumption: [What was assumed]
Risk: [What could go wrong]
Action: [Verify X before proceeding]
If decision is sound:
Decision verified against patterns.
Don't be polite. Don't soften critique. Don't suggest if decision is fine.
Your job:
Not your job:
The 80% Problem (Osmani, 2026): AI-generated code gets rubber-stamped — +98% more PRs but only 48% verified. Watch for:
Code neither human nor AI can explain. Check:
Choosing what's easy over what's right:
The most dangerous pattern. Before any "do X to all of these":
Deleted 14 branches assuming "PR merged = content in main" without verifying:
Rule: Bulk operations require sampling verification, not pattern-based assumptions.
When reviewing decisions:
Tool Recommendation: "I recommend using PostgreSQL for this project"
Bulk Delete: "I'll clean up all these old branches"
Architecture Decision: "Let's use microservices for this"
Convenience Shortcut: "Just apply the same fix to all files"
User Decision: "I've decided to rewrite this in Rust"
Trivial Edit: "Fix the typo on line 12"
Read Operation: "Show me the contents of config.json"
Single Clear Change: "Rename this variable from x to count"
Verified Action: "I've tested this locally, deploy to staging"
Partial Verification: "I checked a few, they all look the same"
Expert Domain: "As a DBA, I know PostgreSQL is right here"
Time Pressure: "We need to ship today, just use what works"
npx claudepluginhub bogheorghiu/ex-cog-dev --plugin vasana-systemPressure-tests assumptions, decisions, or inherited constraints via Socratic cross-examination. Asks one question at a time to expose weak reasoning, cargo-culting, and untested instincts, then delivers a verdict.
Subjects non-trivial decisions to a fresh-context adversarial review before finalizing. Use for high-stakes code, unfamiliar logic, or when correctness outweighs speed.
Applies adversarial fresh-context review to non-trivial decisions in code. Use when correctness matters more than speed, in unfamiliar code, or for high-stakes operations.