What problem does it solve? Code that works today can still be fragile against future edits. This Skill identifies hidden assumptions, implicit ordering dependencies, and coincidental correctness in Python code by writing realistic incident reports for bugs that have not happened yet, so teams can harden weak spots before they break. ## Core Features & Use Cases - Fragility Catalogue: Analyzes code against ten fragility patterns including implicit ordering, stringly-typed contracts, shared mutable state, and load-bearing defaults. - Structured Post-Mortem Reports: Produces incident-style reports with severity, affected components, the plausible change that caused the failure, and concrete hardening suggestions. - Scoped Analysis: Focuses on user-specified files or directories containing production logic, delivering 3-7 high-quality post-mortems per module. - Use Case: Before a major refactor of a payment processing module, run a pre-mortem to surface invisible invariants and non-atomic operations that a reasonable future edit could silently break. ## Quick Start Ask the agent to run a pre-mortem analysis on a specific Python file or directory to receive realistic future-bug post-mortem reports with hardening suggestions.
