Search for counter-evidence to hypotheses before corroborating data.
npx skills add https://github.com/tomwangowa/agent-skills --skill critical-research
Please help me install this Agent Skill. Skill: critical-research Source: https://github.com/tomwangowa/agent-skills/tree/main/critical-research Command: npx skills add https://github.com/tomwangowa/agent-skills --skill critical-research
This Skill combats cognitive biases like confirmation bias by systematically searching for counter-evidence before accepting supporting evidence, ensuring more objective and rigorous conclusions.
Use the critical-research skill to evaluate the hypothesis that 'containerized applications are always more efficient than virtual machines'.
High-intent search queries and answers about installing and using this skill.
A falsification-first approach systematically prioritizes identifying weaknesses, limitations, and contradictions before accepting supporting evidence, ensuring more objective conclusions. It actively seeks counter-evidence to combat confirmation bias.
To fact-check assertions, apply a falsification-first approach to actively seek counter-evidence, identify limitations, and uncover failure modes before gathering supporting evidence. This delivers clear findings on hypothesis support and conflict analysis.
Verify claims by applying falsification-first research to actively seek counter-evidence before corroborating evidence. This rigorously compares competing approaches by identifying limitations, failure modes, and edge cases to eliminate bias.
Yes, falsification-first research requires web search and fetching capabilities to gather and analyze contradictory and supporting data. This environment setup enables rigorous verification of claims and comparison of competing approaches.
Avoid falsification-first research when gathering purely supportive evidence or exploring strictly theoretical concepts without empirical data. This approach is designed for rigorous verification by actively seeking counter-evidence, limitations, and failure modes.
yy1588133Orchestrate parallel tracer workers to generate, rank, and falsify causal hypotheses.
az9713Analyze claims through a four-step steelman and red-team dialectical framework.
NikitaDmitrieffAnalyze long-form content with multiple thinking frameworks to extract structured insights.
SensLiaoSearch for conditional evidence without altering research scope.
