A curated depot of Claude Code plugins — skills, tools, and extensions.

Add the marketplace, then install any plugin:

/plugin marketplace add seanGSISG/claude-depot
/plugin install trmm-expert@claude-depot

For use in Claude.ai or Claude Desktop, download pre-built .skill files from GitHub Releases:

1. Go to the latest release
2. Download the .skill file for the plugin you want
3. Upload it to Claude.ai (Settings > Skills) or Claude Desktop

.skill files are validated ZIP archives containing the skill and all its reference files. They work identically to plugin-installed skills.

Tactical RMM documentation expert — answers questions about architecture, agents, scripting, checks/tasks, alerting, API, settings, reporting, SSO, integrations, and troubleshooting.

What it provides:

• Expert-level answers from the complete TRMM documentation (46 source files)
• 9 consolidated reference files organized by topic domain
• Intelligent routing that loads only the relevant reference(s) per question
• Quick-reference syntax guide for script variables, custom fields, data queries

Triggers when you ask about:

• TRMM architecture, installation, agent deployment
• Scripting (PowerShell/Python/Bash/Deno), script variables, custom fields
• Checks, tasks, automation policies, alerting, webhooks
• REST API, global settings, permissions, management commands
• Reporting and SSO (Enterprise Edition)
• Third-party integrations (Bitdefender, Zammad), SNMP, troubleshooting

AI-powered search and access to 573 Anthropic documentation paths — Claude Code CLI, API, Agent SDK, prompt library, and more.

What it provides:

• Full-text content search and fuzzy path search across 571 locally mirrored documentation files
• Automatic sync via SessionStart hook (clones/pulls from seanGSISG/claude-code-docs)
• Category-aware results with product context (Claude Code CLI, Claude API, Agent SDK, etc.)
• Graceful degradation — works with or without Python 3.9+

Triggers when you ask about:

• Claude Code CLI features (hooks, skills, MCP, memory, plugins, settings)
• Claude API (messages, batches, files, models, admin, multi-language SDKs)
• Agent SDK (Python/TypeScript, sessions, subagents, custom tools)
• Prompt engineering, tool use, streaming, extended thinking, vision
• Any Anthropic platform documentation

/plugin install claude-docs@claude-depot

Guide for creating effective Claude Code skills with templates, validators, and packaging tools.

What it provides:

• 6-step skill creation workflow (understand > plan > init > edit > package > iterate)
• Python scripts for initializing, validating, and packaging skills
• Reference docs on workflow patterns and output patterns
• Progressive disclosure design principles

Triggers when you:

• Ask to create a new skill
• Want to update or improve an existing skill
• Need help structuring a skill's resources

Audit Claude Code rules and CLAUDE.md files for staleness, quality, and drift against the actual codebase.

What it provides:

• Expert knowledge on rules quality, anti-patterns, and progressive disclosure best practices
• Automated discovery and audit of all .claude/rules/*.md and CLAUDE.md files
• Staleness detection: dead globs, broken file references, outdated conventions, convention drift
• Freshness scoring (fresh / needs-review / stale) with actionable recommendations
• /install-rules-audit command to set up GitHub Actions CI integration
• Standalone workflow that audits affected rules on every PR using claude-code-action

Triggers when you ask to:

• Audit or review rules quality and freshness
• Find stale or broken rules
• Improve progressive disclosure setup
• Check if CLAUDE.md files are up to date
• Install rules audit CI workflow

/plugin install rules-auditor@claude-depot

Hooks that enable Claude Code to learn, reflect, and persist knowledge across sessions.

What it provides:

• Automatic loading of prior learnings at session start (injected into Claude's context)
• Silent tool usage tracking during sessions (success and failure patterns)
• End-of-session reflection that detects repeated failure patterns and stores insights
• Two-tier memory: global (~/.claude/learnings.db) cross-project, local (.claude-memory/learnings.db) per-repo
• Automatic decay of stale, low-value learnings (60-day TTL)

Requirements: sqlite3 (pre-installed on macOS/Linux), jq (optional)

/plugin install continual-learning@claude-depot

To add a new plugin, create a subdirectory under plugins/ with a .claude-plugin/plugin.json manifest and add an entry to the root marketplace.json.

See CLAUDE.md for development conventions and the release workflow.

MIT