Mechanism over ritual
Each skill implements the durable cognitive move, named for what it does, not a trademarked brand.
A library of thinking frameworks (premortem, reference-class forecasting, argument mapping, and more) rebuilt for agents. Three things make it different from a list of mental models:
Mechanism over ritual
Each skill implements the durable cognitive move, named for what it does, not a trademarked brand.
Honest evidence grading
Every skill carries a tier (S/M/P/V/A/C/X) and a dossier stating what the research does and does not show. Including when the answer is “the evidence is weaker than people think.”
Artifact, not prose
Every skill produces a concrete, reusable deliverable: a risk register, an option matrix, an argument map, a Thinking Plan.
Explicit 'when not to use'
Each skill says where it misleads, to guard against cargo-cult execution.
I'm not sure where to start
Read the Framework Advisor: describe your situation and get a prioritized plan of which frameworks to use and why.
Teach me by domain
Browse the domains: the families of cognitive operations, from reframing to stress-testing to reflection.
Show me the evidence
The bibliography is the honest-grading promise made browsable: every claim traces to a graded source.
Solve a recurring job
The recipes chain skills end to end: reframe a problem, expand options, stress-test a decision, audit reasoning.