Review project plans, designs, and findings against 10 failure dimensions.
npx skills add https://github.com/victoriacity/openakari --skill critique-victoriacity
Please help me install this Agent Skill. Skill: critique Source: https://github.com/victoriacity/openakari/tree/main/.claude/skills/critique Command: npx skills add https://github.com/victoriacity/openakari --skill critique-victoriacity
This Skill provides an adversarial review of plans, findings, or designs to proactively identify weaknesses, gaps, and potential failure modes before they are committed.
Use the critique skill to review the attached file 'project_plan_v1.md'.
High-intent search queries and answers about installing and using this skill.
Adversarial review of a project plan involves evaluating the document against 10 specific failure dimensions like layer misattribution and scope drift to identify weaknesses and potential failure modes before commitment.
Scope drift is a failure dimension where a project expands beyond its original boundaries. Adversarial critique identifies it by rigorously challenging design assumptions to ensure robustness and correctness against potential failure scenarios.
Yes, you can use this adversarial review process on project plans, design documents, and research findings to proactively identify gaps, weaknesses, and potential failure modes.
You need an adversarial review for your design document before finalizing it, as proactively catching issues early prevents compounding errors and ensures robustness against potential failure scenarios.
Yes, adversarial critique works without additional dependencies, allowing you to directly load artifacts like project plans and design documents to evaluate their robustness and correctness.
galz10Evaluates technical research documents against Documentarian standards and outputs structured Markdown reviews with actionable feedback.
riethmayerIdentify failure modes in plans, prompts, or strategies by reasoning backward from outcomes.
TheSethRoseMap identified threats to security controls with risk scoring and mitigation plans.
iulianditaScore, lint, and refactor AI skill collections through iterative peer review.
