
Claude Skills by tony
github.com/tonyWeave prompt — compare independent adversarial implementations, then pick the best approach
Use when asked to refine a draft, iterate on code until cleaner, or polish a README through repeated critique and improvement.
Weave code review — run independent adversarial reviews in parallel, then synthesize findings
Use when brainstorming and refining developer-experience, documentation, or tooling UX through the Serene DX lenses.
Assess a decision against stated criteria: sourced ratings per criterion and a verdict, kept in a dated topic folder
Pick up a research topic or assessment: carry sources into today's run, re-verify them, then fill gaps or expand scope
Use when researching or assessing a topic and keeping its sources, picking up earlier research, filling gaps, re-checking references, or reporting.
Write a topic's report: the answer and summary first, findings next, then method, exhibits, and sources
List research topics and assessments: latest run, source states, open questions, and the next action for each
Re-check a topic's sources (vanished, moved, changed, misquoted), look for missed sources, and lint the report
Assess a decision against stated criteria: sourced ratings per criterion and a verdict, kept in a dated topic folder
Pick up a research topic or assessment: carry sources into today's run, re-verify them, then fill gaps or expand scope
Use when researching or assessing a topic and keeping its sources, picking up earlier research, filling gaps, re-checking references, or reporting.
Write a topic's report: the answer and summary first, findings next, then method, exhibits, and sources
List research topics and assessments: latest run, source states, open questions, and the next action for each
Re-check a topic's sources (vanished, moved, changed, misquoted), look for missed sources, and lint the report