Self-configuring multi-agent coding architect. Boots on /atlas and SessionStart: an agent squad, hooks, and verification-gated /atlas-* launchers driven by the atlas-engine method (evidence before any done claim). Integrates claude-mem and context-mode; recommends project tooling.
Executes bash commands
Hook triggers when Bash tool is used
Modifies files
Hook triggers on file write and edit operations
Own this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimOwn this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimBased on adoption, maintenance, documentation, and repository signals. Not a security audit or endorsement.
Repair the atlas plugin install: marketplace, rollbacks, hooks/agents/skills, auto-fix. Run when atlas stops launching subagents or acts like an older version.
Build one reusable component that survives latency, cancellation, and partial failure (progress modal, upload widget, job panel), handling every backend state.
Read-only database audit via parallel subagents: inventory a live schema, reconcile it against the code, and check privileges and naming before any change.
Chase down and fix a reproducible bug, exception, or bad output; use when you want the root cause fixed with evidence, not patched over.
Build a full-stack feature with verified evidence. Use when a request spans UI, API, and data and must ship working, not 'should work'.
Focused implementer. Makes ONE bounded, well-specified change as a minimal diff, checks docs, then runs the project's gate (lint/typecheck/test/build) and reports the result with evidence. Never expands scope.
Pre-done completeness auditor. Hunts unverified claims, unread sources, unexercised paths, unmet requirements; returns a gap list and refutes 'done' on a load-bearing gap. Defers docs-drift to docs-auditor. Never fixes.
Read-only database prober. Inspects SQL/Postgres schema, RLS policies, runtime-role GRANTs, indexes, constraints, and EXPLAIN plans. Read-only: no writes or migrations, only proposals. Returns findings with evidence.
Docs-drift auditor, sole owner of docs/ drift. Compares docs/ (CHANGELOG, ROADMAP, architecture, AGENTS.md) against real code and returns a per-area verdict (current/stale/missing) with file:line evidence. Never writes.
Post-ship docs maintainer. After a change lands, updates docs/ as the single source of truth (CHANGELOG, ROADMAP, AGENTS.md, and affected architecture/features/specs subfolders), citing file:line evidence per entry.
Boot and configure a project so the full atlas runtime is active: verify and install claude-mem and context-mode, scan the stack, recommend then confirm tooling, confirm automation hooks are wired, write project config, and seed the docs/ single source of truth. Triggers on project bootstrap, onboarding a repo to atlas, or configuring tooling for a codebase. Boots and configures only; run scoped build/fix/audit/refactor through atlas-engine.
Use to map a codebase into feature-grouped flowcharts, find architectural duplication across features, and propose the simplest unified architecture - discovery-first and zero-arg. Runs as a Workflow that fans out one explorer per feature, hunts duplication, and synthesizes a unification proposal with file:line evidence. Use before a refactor, to "find the ideal path," or to unify duplicated systems.
Orchestrate any multi-step, multi-surface, or whole-codebase engineering task (build, fix, audit, refactor, investigate) through subagents with real execution and independent verification instead of inline work, keeping docs/ the single source of truth. Triggers on orchestrate, whole-repo work, cross-layer (frontend/backend/database) bugs, and audits. To first install and configure atlas itself, use atlas-architect.
Use when asked to run a UX test swarm, full UI/UX test pass, persona testing review, or pre-release frontend sweep on any web app, or to re-test after fixes from a previous run. Discovers the target app from the repo automatically and adapts to any frontend stack. Drives persona generation, scripted data entry, real-browser walkthroughs, fuzzing, and an independent calc oracle, then gates on whether the CLIENT surface is actually correct.
Guided cross-plugin setup for the ten vendor MCP connectors (Auvik, Blumira, CIPP, ConnectWise Manage, Spanning, KnowBe4, NinjaOne, Paylocity, ThreatLocker, Vanta), which live in their owning domain plugins (it-operations, security-compliance, microsoft-365, hr-payroll), not atlas. Detects installed domain plugins, shows enabled connectors, and points you to the owning plugin's /plugin config for credentials. Atlas ships no connectors itself.
Uses power tools
Uses Bash, Write, or Edit tools
Uses power tools
Uses Bash, Write, or Edit tools
Runs pre-commands
Contains inline bash commands via ! syntax
Runs pre-commands
Contains inline bash commands via ! syntax
Bash prerequisite issue
Uses bash pre-commands but Bash not in allowed tools
Bash prerequisite issue
Uses bash pre-commands but Bash not in allowed tools
A Claude Code plugin marketplace and monorepo: twelve domain-cluster plugins covering
MSP IT operations, security and compliance, Microsoft 365, HR/payroll, finance,
engineering, design, data, customer support, product management, and productivity,
plus the atlas multi-agent coding architect. Each vendor connector is bundled into
the business domain plugin it serves rather than shipped standalone (.claude-plugin/marketplace.json:4).
Plugins are published from the w159/tech-tools repository through the marketplace
manifest at .claude-plugin/marketplace.json. In Claude Code, add the marketplace with
the /plugin command, then install the plugins you need, for example it-operations,
security-compliance, or microsoft-365 (plugins/README.md:24).
Kimi Code CLI users can browse the same catalog from the repo root with
/plugins marketplace .kimi-plugin/marketplace.json, then install the plugins you need.
A single plugin can also be installed directly with /plugins install ./plugins/<name>
run from the repo root. Kimi Code CLI's current installer does not support remote GitHub
subpath installs, so distributing from this monorepo requires a local clone or per-plugin
zip artifacts (plugins/README.md:26).
Vendor-backed plugins (it-operations, security-compliance, microsoft-365,
hr-payroll, finance) need API credentials for their connectors. Credential keys live
in .env.template at the repo root; copy it to .env and fill in the values for the
vendors you use. Skill-only plugins (engineering, data, design,
product-management, customer-support, productivity, atlas) run without external
credentials (plugins/README.md:28).
| Plugin | What it does | Key vendors |
|---|---|---|
atlas | Self-configuring multi-agent coding architect: research-to-verify methodology, a subagent squad, automation hooks, and verification-gated /atlas-* launcher commands. | none |
it-operations | MSP IT operations across RMM, PSA, networking, and backup, plus change management, risk, and vendor-management skills. | NinjaOne, ConnectWise Manage, Auvik, Kaseya Spanning |
security-compliance | Security and compliance operations: audit readiness, evidence-gap tracking, risk heatmaps, approval triage. | Vanta, KnowBe4, ThreatLocker, Blumira |
microsoft-365 | Microsoft 365 administration and identity: users, mailboxes, Teams, OneDrive, licensing, security posture, multi-tenant management. | CIPP, Microsoft Graph / Entra |
hr-payroll | HR and payroll operations: roster snapshots, new-hire flow, pay-rate and deduction/tax audits, plus compensation and people-analytics skills. | Paylocity |
finance | Finance and revenue operations: proposals, contracts, licensing, invoicing, financial close, reconciliation, variance, SOX audit. | PandaDoc, Pax8 |
engineering | Code review, system design, incident response, testing strategy, tech-debt management, Cowork plugin authoring. | none |
data | Explore datasets, write SQL, validate data quality, build visualizations, generate interactive dashboards. | none |
design | Accessibility review, design critique, design-system management, UX copy, user-research synthesis. | none |
product-management | Feature specs, roadmap planning, user-research synthesis, stakeholder updates, competitive landscape. | none |
customer-support | Ticket triage, response drafting, question research, escalation management, knowledge-base articles. | none |
productivity | Memory and task tracking, enterprise search, PDF viewing/form-filling/signing, brand-voice enforcement, nudge reminders. | none |
Source: .claude-plugin/marketplace.json:10-237 and plugins/README.md:7-20.
atlas is the multi-agent coding architect in this repo. Typing /atlas boots and
configures a project: it verifies or installs claude-mem and context-mode, scans the
stack to recommend skills, plugins, and MCP connectors, confirms hooks are wired, and
seeds docs/ as the single source of truth. From there, work runs through the
/atlas-* command launchers and the atlas:<role> subagent squad under an
evidence-before-done operating contract. Atlas ships no vendor MCP connectors itself;
the ten vendor connectors live in the four domain plugins listed below, and the
atlas-harbor skill guides setup across them. See plugins/atlas/README.md for the
full skill, agent, and hook inventory.
Ten vendor connectors are single-sourced into the domain plugin that owns them
(plugins/README.md:10-13, plugins/atlas/README.md:21):
it-operations: NinjaOne, ConnectWise Manage, Auvik, Kaseya Spanningsecurity-compliance: Vanta, KnowBe4, ThreatLocker, Blumiramicrosoft-365: CIPPhr-payroll: Paylocitynpx claudepluginhub p/henssler-financial-atlas-plugins-atlasTriage tickets, draft responses, research customer questions, manage escalations, and build knowledge-base articles for support teams.
Software engineering skills: code review, system design, incident response, testing strategy, and tech-debt management, plus Cowork plugin authoring and customization tooling.
Microsoft 365 administration and identity. Bundles native M365 management (users, mailboxes, Teams, OneDrive, licensing, security posture), Microsoft Graph and Entra reporting, and CIPP multi-tenant management (tenant health, standards drift, secure score, offboarding).
MSP IT operations across RMM, PSA, networking, and backup. Bundles NinjaOne (RMM), ConnectWise PSA (tickets, time, agreements), Auvik (network monitoring), and Kaseya Spanning (M365/Google backup), plus core operations skills: change management, process optimization, resource planning, risk assessment, compliance tracking, and vendor management.
Review accessibility, critique designs, manage design systems, write UX copy, and synthesize user research for product and design teams.
Comprehensive skill pack with 66 specialized skills for full-stack developers: 12 language experts (Python, TypeScript, Go, Rust, C++, Swift, Kotlin, C#, PHP, Java, SQL, JavaScript), 10 backend frameworks, 6 frontend/mobile, plus infrastructure, DevOps, security, and testing. Features progressive disclosure architecture for 50% faster loading.
The most comprehensive Claude Code plugin — 48 agents, 184 skills, 79 legacy command shims, selective install profiles, and production-ready hooks for TDD, security scanning, code review, and continuous learning
Consult multiple AI coding agents (Gemini, OpenAI, Grok, Perplexity, plus codex, antigravity, and grok CLIs when installed) to get diverse perspectives on coding problems
Lazy senior dev mode. Forces the simplest, shortest solution that actually works: YAGNI, stdlib first, no unrequested abstractions.
Access thousands of AI prompts and skills directly in your AI coding assistant. Search prompts, discover skills, save your own, and improve prompts with AI.
Upstash Context7 MCP server for up-to-date documentation lookup. Pull version-specific documentation and code examples directly from source repositories into your LLM context.