Agent Skills by Nicholas Stelter
Showing 23 vetted skills indexed across 1 GitHub repositories.
steltzaudit-ui-handlers
Audit a single page for unfinished UI handlers via static analysis.
steltzanalyze-overfetch
Inventory GraphQL queries and fragments, trace field usage, and generate audit-ready optimization plans.
steltzsft-versioning-guide
Automate semantic versioning and changelog updates for skill-factory plugins.
steltzsft-cross-cutting-patterns
Enforce standardized SKILL.md frontmatter and section conventions across skills.
steltzsft-coordination-patterns
Select multi-agent coordination patterns for Claude Code skill workflows.
steltzsft-scaffold-plugin
Generate plugin directory structures with plugin.json and SKILL.md templates.
steltzsft-skill-anatomy
Analyze Claude Code skill structure and section rationale for audits.
steltzsft-build-plugin
Automate the brainstorm, plan, implement, and version workflow for skills and plugins.
steltzsft-skill-comparison-matrix
Compare structural templates for Claude Code skill design.
steltzusing-git-worktrees
Create isolated git worktrees sharing the same repository for feature work.
steltztest-driven-development
Enforce test-driven development with a red-green-refactor workflow.
steltzsystematic-debugging
Enforce a four-phase systematic debugging process to identify and fix root causes.
steltzusing-superpowers
Invoke the Skill tool before responding to load skill directives.
steltzdispatching-parallel-agents
Dispatch dedicated agents to investigate independent failures in parallel.
steltzexecuting-plans
Execute implementation plans with critical review and mandatory verifications.
steltzfinishing-a-development-branch
Automate feature branch finishing with merge, PR, keep, or discard options.
steltzbrainstorming
Clarify goals and constraints to produce concrete design outlines.
steltzwriting-plans
Generate implementation plans from formal specs with file mapping and task decomposition.
steltzrequesting-code-review
Automate code-review planning and validation with standardized templates and diff-based assessment.
steltzreceiving-code-review
Evaluate code-review feedback through a six-step verification workflow before implementing changes.
steltzwriting-skills
Codify frontmatter and structured guidance for CLAUDE Skill Units.
steltzverification-before-completion
Run verification commands and collect fresh evidence for completion claims.
steltzsubagent-driven-development
Dispatch fresh subagents for each plan task with two-stage reviews.