Agent Skills by Holden Ramsey
Showing 18 vetted skills indexed across 1 GitHub repositories.
agent-creator
Generate custom Task subagent definitions in the ~/.claude/agents/ directory.
using-git-worktrees
Create isolated git worktrees with safety checks and project setup.
systematic-debugging
Diagnose software bugs through a four-phase root-cause investigation framework.
incremental-implementation
Decompose multi-day features into independently deployable increments with migration plans.
verification-before-completion
Enforce fresh verification evidence before claiming task completion.
Example Skill
Enforce a three-phase workflow with checklists and anti-pattern guidance.
skill-creator
Generate standardized SKILL.md frontmatter and resource scaffolds for Claude Skills.
command-creator
Create slash command skeletons in designated commands directories using a Markdown template.
receiving-code-review
Evaluate external code review feedback for technical correctness and verifiability.
using-tmux
Automate interactive CLI tools via tmux send-keys in detached sessions.
test-driven-development
Enforce test-first development with failing tests before production code.
subagent-driven-development
Dispatch fresh subagents per task with post-task code reviews.
memory-management-optimization
Debug memory leaks and optimize allocations in C++ and Python.
batch-operations
Executes batched identical operations across files or Git repositories with parallel tool calls and validation checks.
code-smell-detection
Detect code smells and output evidence, locations, severity, and refactoring recipes.
hook-creator
Create native Claude Code hook scripts with JSON output and templates.
Processes PDFs with text/table extraction, merging, splitting, rotation, OCR, and form filling.
mcp-cli
Explore MCP server tools, resources, and prompts via CLI.