17 agent skills (SKILL.md, the open Agent Skills standard): engineering discipline, research, safe process termination, and rollback thinking. Installable in Claude Code / OpenCode / Codex / Gemini / any harness.
aidvizhhub/caveman-dvizh-skills
Skill Claude CodeCodex
A guided way to review completed work and suggest several possible next steps, including improvements, new ideas, risks, and industry-style options.
aidvizhhub/caveman-dvizh-skills
Skill Claude CodeCodex
A problem-solving approach that looks for shortcuts, leverage points, and alternative assumptions instead of attacking a task directly.
aidvizhhub/caveman-dvizh-skills
Skill Claude CodeCodex
A coding method that starts by studying official examples and existing high-quality code before writing something new.
aidvizhhub/caveman-dvizh-skills
Skill Claude CodeCodex
A response-writing rule that requires every answer to end with a brief conclusion and likely follow-up questions with their answers.
aidvizhhub/caveman-dvizh-skills
Skill Claude CodeCodex
A guide for starting programs automatically and handling desktop tasks such as OBS frame-rate setup, Linux game launching, and video editing.
aidvizhhub/caveman-dvizh-skills
Skill Claude CodeCodex
A troubleshooting guide that starts with local evidence such as logs and system state, then checks relevant web information before changing one thing at a time.
aidvizhhub/caveman-dvizh-skills
Skill Claude CodeCodex
A safety guide for stopping unwanted computer processes without accidentally stopping the command used to find them.
aidvizhhub/caveman-dvizh-skills
Skill Claude CodeCodex
A step-by-step workflow for taking a software task from requirements and planning through implementation, verification, and a final report.
aidvizhhub/caveman-dvizh-skills
Skill Claude CodeCodex
A risk-prevention workflow that looks for likely failures before work begins and plans ways to detect or avoid them.
aidvizhhub/caveman-dvizh-skills
Skill Claude CodeCodex
A research workflow that gathers information from multiple sources, including documentation, web searches, and industry code examples.
aidvizhhub/caveman-dvizh-skills
Skill Claude CodeCodex
A set of operating rules for running long-lived background processes and detecting when they finish or stop.
aidvizhhub/caveman-dvizh-skills
Skill Claude CodeCodex
A set of rules for keeping coding-agent work small, direct, and resource-conscious. It encourages batching work, limiting file and function size, checking ideas before adding code, and verifying results.
aidvizhhub/caveman-dvizh-skills
Skill Claude CodeCodex
A conversation style that makes the agent speak like a caveman and follow a tribal rulebook. It also describes a broader system of skills for planning, researching, checking, coding, and remembering lessons.
aidvizhhub/caveman-dvizh-skills
Skill Claude CodeCodex
A teaching guide for coding agents that adjusts explanations to the learner’s current understanding, moving from concrete examples to broader ideas.
aidvizhhub/caveman-dvizh-skills
Skill Claude CodeCodex
A problem-solving guide for coding agents that explores possible approaches, checks a draft against set criteria, and can return to an earlier approach when a path fails.
aidvizhhub/caveman-dvizh-skills
Skill Claude CodeCodex
A verification workflow for checking the current state of work against documented requirements or a plan at each step.
At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: