Score technical implementation options across 8 dimensions to trigger Gate 4 review.
npx skills add https://github.com/WellApp-ai/Well --skill tech-divergence
Please help me install this Agent Skill. Skill: tech-divergence Source: https://github.com/WellApp-ai/Well/tree/main/cursor-rules/skills/tech-divergence Command: npx skills add https://github.com/WellApp-ai/Well --skill tech-divergence
This Skill provides a disciplined framework to evaluate multiple technical options for a given problem by applying an 8-dimension scoring matrix and signaling when Gate 4 should involve human decision-makers.
Apply tech-divergence during Plan Mode Phase 2 to score options A, B, and C and determine whether to auto-proceed or escalate to Gate 4 for review.
High-intent search queries and answers about installing and using this skill.
To evaluate technical architectures, apply an 8-dimension scoring matrix that scores patterns, scope, data models, dependencies, API surface, reversibility, security, and performance to output a recommended approach or flag for human review.
Human review for architectural decision-making is triggered at Gate 4 when high-stakes technical divergence is detected, whereas low-risk scores auto-proceed without escalation to human decision-makers.
Scoring data models and dependencies during technical divergence involves applying a structured scoring matrix across 8 dimensions to compare alternative architectures and determine whether to auto-proceed or escalate to Gate 4.
The best way to compare performance implications across valid technical approaches is using a structured scoring matrix during Plan Mode Phase 2 to quantify options and output a recommended approach or flag for Gate 4 human review.
Yes, the architecture evaluation process features auto-proceed behavior for low-risk scores, automatically continuing without human input and only triggering Gate 4 for high-stakes architectural decisions that require manual review.
The scoring matrix is specifically designed for Plan Mode Phase 2 (Technical Diverge) to compare architectures, data models, dependencies, and performance implications across multiple valid approaches to determine Gate 4 triggers.
mpazarynaAnalyze technical options and draft an Architecture Decision Record.
thejefflarsonGate LLM outputs by confidence and route high-stakes cases to human review.
pradyumnasagarAutomate inter-skill data transfer and provenance capture via JSON Material Passports.
asd335w3dCompare technical solutions with weighted scoring and output a comparison matrix.
