| name | pauli-mythos-reasoning |
| description | Reasoning harness for context normalization, expert routing, bounded iteration, and decision summaries without chain-of-thought leakage. |
| version | 1.0.0 |
| required_environment_variables | [] |
Pauli Mythos Reasoning
triggers
- reasoning
- evaluate options
- council
- confidence
- synthesize
when_to_use
Use when the task needs structured decision synthesis across multiple candidate approaches.
when_not_to_use
Do not use when a simple direct answer or straightforward code change is enough.
required_tools
required_env
context_budget
- one reasoning skill plus domain skill companions
safety_gates
- do not expose hidden chain-of-thought
- produce decision summaries and trace notes only
workflow
- Normalize the problem and constraints.
- Compare options with bounded loops.
- Emit a clear decision record and confidence note.
output_contract
- normalized problem statement
- compared options
- decision
- confidence
tests
- output excludes chain-of-thought
- final recommendation is concrete