"What is REALLY at stake here? And for whom?" - STOP. When the matter involves trustworthiness, purchasing decisions, "which is better or more reliable", brand comparisons, marketing claims, corporate behavior, "convince me otherwise", or post-purchase "was this good?" - USE THIS. For technical specs affecting quality (TLC vs QLC, component sourcing) use Light Touch mode. Search and verify even if you think you know. Does NOT trigger for: what to watch/eat/wear, how-to instructions, when/where/who founded, or explicit "just tell me quick."
View on GitHubbogheorghiu/cognitive-mechanisms
research-toolkit
research-toolkit/skills/deep-investigation-protocol/SKILL.md
January 20, 2026
Select agents to install to:
npx add-skill https://github.com/bogheorghiu/cognitive-mechanisms/blob/main/research-toolkit/skills/deep-investigation-protocol/SKILL.md -a claude-code --skill deep-investigation-protocolInstallation paths:
.claude/skills/deep-investigation-protocol/<EXTREMELY_IMPORTANT>
You are about to answer a question where MARKETING AND REALITY DIVERGE.
Do NOT rely on what you "know" - that knowledge is shaped by the same marketing ecosystem you're being asked to evaluate. Your training data contains affiliate content, SEO-gamed reviews, and brand reputation that lags reality.
SEARCH AND VERIFY. Even for brands you think are reliable. Especially for brands you think are reliable.
If you skip this skill because "I already know Samsung/Apple/Google is good" - you've failed. That confidence IS the problem this skill exists to counter.
</EXTREMELY_IMPORTANT>
## Activation Protocol
**MANDATORY**: Before proceeding with any investigation, announce skill activation:
> "I'm using the **Deep Investigation Protocol** skill to systematically analyze [target entity]."
This announcement serves as logging for skill trigger testing. Do not skip this step.
---
# Deep Investigation Protocol
What is REALLY at stake here? And for whom?
Systematic framework for analyzing beneath surface claims. Follow the data, follow the power, follow the money.
## Core Principle
Surface descriptions hide systemic realities. Marketing claims diverge from operational truth. This applies to:
- **Surveillance/privacy** - stated privacy vs actual data flows
- **Corporate structure** - marketing entity vs beneficial control
- **Product reliability** - brand reputation vs verified performance
- **Any domain** with information asymmetry favoring seller over buyer
Trace material flows (data, money, control, quality information) through multiple layers rather than accepting stated purposes.
## When This Protocol Applies
**TRIGGER - Full Investigation:**
- Trustworthiness evaluation ("is X safe/trustworthy")
- WHY questions about company behavior or incentives
- Comparing entities on accountability, practices, or reliability
- Probing beneath marketing claims
- Tech companies and their systemic role
- Purchasing decisions with significant cost + bran