claude-system-management
Create, update, and audit .claude/ system configuration including agents, skills, memory, rules, hooks, and CLAUDE.md
Skill by reshinto
Install
Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.
npx agentmarket install skill/reshinto-claude-system-management
Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.
Get the weekly diff
The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.
About
Use cases: security
Reviews
Write a reviewNo reviews yet. Installed it? Reviews come from verified installers only. How reviews work
Permissions
What this skill can do and what it asks for, from its SKILL.md. Checked Oct 25, 2026.
- Tools
- No tools found
- Secrets
- None asked for
- Install scripts
- None found
Network hosts it talks to 1
codeload.github.com· found in install command
Files it touches
~/.claude/skills/claude-system-management
What's inside
- Skill name
claude-system-management- Description
- Create, update, and audit .claude/ system configuration including agents, skills, memory, rules, hooks, and CLAUDE.md
- License
- MIT
- Instructions
- 103 lines in SKILL.md · view on GitHub
Files (1)
- SKILL.md
SKILL.md (first lines)
# Claude System Management ## Task Create, update, or audit `.claude/` system configuration to ensure all components are consistent, current, and aligned with the project's codebase. ## Instructions 1. Identify the target: specific component (agent, skill, rule, hook) or full system audit 2. Verify against the standards below 3. Fix or report inconsistencies ## Component Standards ### Agent Definitions (`.claude/agents/*.md`) Frontmatter schema (all keys required): ```yaml --- name: <lowercase-hyphenated> description: <one-line summary> tools: [Read, Glob, Grep] or [Bash, Read, Glob, Grep] model: sonnet maxTurns: <8-15> --- ``` Content structure: 1. `# Agent Name` — title 2. `## Role` — one-paragraph purpose 3. `## Review Areas` — numbered checklist of responsibilities 4. `## Required Skills` — bullet list of domain expertise 5. `## Constraints` — bullet list of rules the agent must follow 6. `## Output Format` — labeled format (PASS/FAIL, APPROVED/BLOCKED, etc.) ### Skill Definitions (`.claude/skills/<name>/SKILL.md`) Frontmatter schema: ```yaml --- name: <lowercase-hyphenated> description: <one-line summary> user-invocable: true --- ``` Content structure: 1. `# Skill Name` — title 2. `## Task` — one-paragraph purpose 3. `## Instructions` — numbered steps 4. Skill-specific sections (checklists, review areas, steps) 5. `## Output Format` — if applicable ### Memory (`.claude/projects/<path>/memory/`) - `MEMORY.md` index stays under 200 lines - Each memory file has frontmatter: name, description, type (user/feedback/project/reference) - No duplicate memories — update existing before creating new - Stale memories pruned when discovered ### Rules (`.claude/rules/*.md`) - Each rule file covers one topic - No contradictions between rule files - Rules align with CLAUDE.md summary - Project-specific conventions, not generic best practices ### CLAUDE.md - Reflects current tech stack and key paths - Concise — detailed rules in `.claude/rules/`, not here - No outdated references to removed features or changed conventions ## CLAUDE.md Quality Audit - **Quality scoring**: Rate CLAUDE.md sections for accuracy, completeness, and token efficiency - **Template adherence**: Verify structure matches the project's established format (Tech Stack, Architecture, Key Paths, Rules, Testing, Workflow) - **Targeted improvements**: Suggest specific edits to improve signal density without increasing length - **Drift detection**: Compare CLAUDE.md claims against actual codebase state ## Audit Procedure When running a full audit: 1. List all agents, verify frontmatter + content structure 2. List all skills, verify frontmatter + content structure 3. Cross-check: agents reference skills that exist, workflow references agents that exist 4. Verify PLAN.md directory listing matches actual files 5. Verify CLAUDE.md key paths exist in the filesystem 6. Check rules for contradictions with each other and with CLAUDE.md 7. Score CLAUDE.md quality and suggest improvements ## Output Format - VALID: [component] - configuration is correct and current - DRIFT: [component] - mismatch between config and codebase + fix - STALE: [component] - outdated entry to update or remove
From reshinto/dev-forge, MIT.
More from reshinto/dev-forge
- 0.406score
strict-type-review
reshinto
Review code for strict type safety compliance including checked indexed access, discriminated unions, tuple types, and runtime validation
SkillDFreshness: DClaude Code +9 - 0.406score
debugging
reshinto
Systematic debugging with diagnostic paths covering data processing, rendering, state machine, asset loading, form/input, and build issues
SkillDFreshness: DClaude Code +9 - 0.406score
verification
reshinto
Pre-completion verification with feature completeness checks, coverage validation, branch safety, and state management review
SkillDFreshness: DClaude Code +9 - 0.406score
readme-optimization
reshinto
Optimize README and public-facing content for GitHub discoverability, clear presentation flow, and accurate feature positioning
SkillDFreshness: DClaude Code +9 - 0.406score
feature-dev
reshinto
Guided feature development following a 7-step workflow covering product evaluation, design, architecture, implementation, review, QA, and documentation
SkillDFreshness: DClaude Code +9 - 0.404score
accessibility-audit
reshinto
Audit UI components for WCAG 2.1 AA compliance, animation accessibility, and design token consistency
SkillDFreshness: DClaude Code +9 - 0.404score
branch-safety-check
reshinto
Verify the current git branch is a unique task-related working branch and create one if needed
SkillDFreshness: DClaude Code +9 - 0.404score
architecture-review
reshinto
Evaluate architectural decisions for state management design, build optimization, and security-by-design patterns
SkillDFreshness: DClaude Code +9 - 0.404score
documentation-review
reshinto
Review documentation for clarity, structural consistency, accuracy, and contributor onboarding quality
SkillDFreshness: DClaude Code +9 - 0.404score
implementation-planning
reshinto
Plan implementation phases for new features following a structured phase, file, test, and dependency template
SkillDFreshness: DClaude Code +9 - 0.404score
security-coverage-audit
reshinto
Run security checks (OWASP client-side) and verify test coverage thresholds with E2E validation
SkillDFreshness: DClaude Code +9 - 0.404score
tdd
reshinto
TDD workflow using a 4-part test matrix (unit/correctness, integration, component/story, E2E)
SkillDFreshness: DClaude Code +9
Signals
- Stars
- 0
- Forks
- 0
- Open issues
- 0
- Last commit (this folder)
- Apr 8, 2026 (6 months ago)
From GitHub (fetched Oct 24, 2026).
Details
- License
- MIT
- Repository
- github.com/reshinto/dev-forge
- First seen
- Oct 24, 2026
- Last checked
- Oct 24, 2026
- Source
- GitHubgithub-graphql
More from reshinto
- 0.406score
strict-type-review
reshinto
Review code for strict type safety compliance including checked indexed access, discriminated unions, tuple types, and runtime validation
SkillDFreshness: DClaude Code +9 - 0.406score
debugging
reshinto
Systematic debugging with diagnostic paths covering data processing, rendering, state machine, asset loading, form/input, and build issues
SkillDFreshness: DClaude Code +9 - 0.406score
verification
reshinto
Pre-completion verification with feature completeness checks, coverage validation, branch safety, and state management review
SkillDFreshness: DClaude Code +9 - 0.406score
readme-optimization
reshinto
Optimize README and public-facing content for GitHub discoverability, clear presentation flow, and accurate feature positioning
SkillDFreshness: DClaude Code +9 - 0.406score
feature-dev
reshinto
Guided feature development following a 7-step workflow covering product evaluation, design, architecture, implementation, review, QA, and documentation
SkillDFreshness: DClaude Code +9 - 0.404score
accessibility-audit
reshinto
Audit UI components for WCAG 2.1 AA compliance, animation accessibility, and design token consistency
SkillDFreshness: DClaude Code +9
Cite this page
claude-system-management skill. Agent Mart. https://agentmart.dev/skills/reshinto-claude-system-management (last updated 2026-10-24)