Agent Skills by sjw2005
Showing 22 vetted skills indexed across 1 GitHub repositories.
planning-with-files
Create and maintain task_plan.md, findings.md, and progress.md files for multi-step tasks.
using-git-worktrees
Create isolated Git worktree workspaces with automatic directory selection and safety checks.
test-driven-development
Enforce failing tests before coding with a Red-Green-Refactor workflow.
systematic-debugging
Locate root causes of bugs through four systematic debugging phases.
pi-planning-with-files
Organize complex tasks into persistent planning files with automatic session recovery.
using-superpowers
Enforces Skill tool invocation before answering to ensure governed conversations and task routing.
dispatching-parallel-agents
Dispatch independent tasks to specialized agents with isolated contexts.
obsidian-dataview
Extract and analyze Obsidian note metadata using Dataview queries and DataviewJS.
find-skills
Discover and install agent skills using the Skills CLI.
executing-plans
Load, review, and execute written plans step-by-step with verifications.
finishing-a-development-branch
Finalize a development branch with merge, PR, or cleanup options.
obsidian-excalidraw
Generate Obsidian-integrated Excalidraw v2 JSON diagrams with wiki links.
self-improvement
Log learnings, errors, and corrections to markdown files under .learnings.
brainstorming
Analyze project goals to surface user intent and design constraints.
templater
Automate Obsidian templating workflows with dynamic content and frontmatter manipulation.
writing-plans
Convert vague project specifications into detailed implementation plans with tasks, milestones, and testing criteria.
requesting-code-review
Automate code review requests with structured templates and SHA tracking.
receiving-code-review
Verify code-review feedback against the codebase before implementing changes.
writing-skills
Create and maintain structured SKILL.md documentation with test-driven workflows.
obsidian-tasks
Manage and query Obsidian Tasks plugin entries with flexible filters, sorts, and groups.
verification-before-completion
Run verification commands and validate outputs before declaring success.
subagent-driven-development
Dispatch a subagent per task and run two-stage reviews for spec compliance and code quality.