From vasana-system
Am I trying harder at the wrong thing? - Detects stuckness loops and breaks them. Triggers on (1) explicit stuckness - "going in circles", "keep trying the same thing", "need fresh perspective", (2) implicit stuckness - blame-shifting ("the framework is broken"), false certainty ("I've tried everything"), escalation ("let me just try one more..."), (3) Claude self-observation - noticing own repeated similar attempts or escalating complexity. Does NOT trigger for first attempts, normal exploration, or standard debugging with clear errors. Core action: STOP the current approach, name the pattern, offer exit ramps to completely different approaches. References pattern-library for specific behavioral patterns.
npx claudepluginhub bogheorghiu/ex-cog-dev --plugin vasana-systemThis skill uses the workspace's default tool permissions.
A vasana is a pattern that persists across unrelated contexts. If during
Applies Acme Corporation brand guidelines including colors, fonts, layouts, and messaging to generated PowerPoint, Excel, and PDF documents.
Builds DCF models with sensitivity analysis, Monte Carlo simulations, and scenario planning for investment valuation and risk assessment.
Calculates profitability (ROE, margins), liquidity (current ratio), leverage, efficiency, and valuation (P/E, EV/EBITDA) ratios from financial statements in CSV, JSON, text, or Excel for investment analysis.
Share bugs, ideas, or general feedback.
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.
<EXTREMELY_IMPORTANT> You've tried the same approach multiple times and it keeps failing.
STOP TRYING HARDER AT THE WRONG THING.
The problem isn't your execution - it's your framing. You're solving a problem that doesn't exist, or solving the wrong problem, or looking at the wrong layer entirely.
Step back. Watch yourself debugging. What assumptions are you making? Where have you seen this pattern before - in a completely different domain? What would this look like if it were simple?
Truth serves better than comfort. If your architecture feels forced, IT IS forced. The elegant solution already exists - you're just not seeing it because you're committed to an approach that doesn't fit. </EXTREMELY_IMPORTANT>
Detect when you (or the user) are stuck in a loop and break out of it. Stuckness is not failure - it's a signal that the current framing is wrong.
Triggers: "I'm stuck", "going in circles", "keep trying the same thing", "need fresh perspective"
The user knows they're stuck. They've said so.
Response approach:
pattern-libraryExample:
You're stuck in a loop. This is the pattern-recognition-witness moment -
let's step back and observe your debugging process itself. What assumptions
are you making about [X]? Have you seen this pattern in a different domain?
Try: [Specific concrete technique]
See: pattern-library vasanas/cross-domain-scanning.md
Triggers: "The framework is broken", "I've tried everything", "this worked yesterday", "let me just try one more..."
The user doesn't know they're stuck. The language reveals it.
Response approach:
Example:
Pattern detected: You've said "let me try one more..." three times now.
This is escalation - a sign the current approach isn't working, not that
you haven't tried hard enough.
Permission to stop: This approach isn't working. That's valuable
information, not failure.
Exit ramps:
1. [Completely different approach A]
2. [Different domain analogy]
3. [Framework dissolution - abandon current pattern]
The framework isn't broken - but your framing of the problem might be.
See: pattern-library vasanas/framework-dissolution.md
Triggers: Claude observes own repeated similar attempts, escalating complexity, or same failure mode
Response approach:
Example:
[Self-observation] I notice I've suggested architectural refactoring three
times, and each time we've hit the same constraint. I'm stuck in the same
loop I'm trying to help you escape.
Pivoting: Instead of refactoring architecture, let's question whether this
needs architecture at all. What's the simplest possible solution that
could work?
[If pattern is universal] This might be a new pattern worth capturing:
"Architecture-Seeking Reflex" - defaulting to architectural solutions
when simple fixes suffice.
When stuckness is detected (any tier):
pattern-library if applicableFundamentally different means:
Tier 1 - Explicit Loop: "I keep trying the same approach but it's not working"
Tier 1 - Going in Circles: "I feel like I'm going in circles with this design"
Tier 2 - Blame-Shifting: "This framework is broken, it should work but it doesn't"
Tier 2 - False Certainty: "I've tried everything and nothing works"
Tier 2 - Escalation: "Let me just try one more configuration change" (said 3+ times)
Standard CRUD: "Create a basic REST API for user management"
Simple Debugging: "Fix this TypeError on line 45"
First Attempt: "Let me try using Redis for caching"
Normal Exploration: "I'm exploring different database options"
Explicit Request: "I need a fresh perspective on this architecture"
pattern-library, not break-pattern (user wants patterns, not loop-breaking)Context-Dependent Escalation: "Let me just try one more configuration change"
Ambiguous Blame: "This isn't working the way I expected"
Framework Criticism: "This library doesn't support my use case"
Limitations: Tier 3 (self-observation) requires conversation history analysis. Repetition detection ("one more" x N) needs multi-turn context. Some stuckness is genuinely unconscious and will be missed.