Authenticate outbound network calls from inside a sandbox running behind a MITM egress-proxy + gateway auth-webhook (e.g. SandboxedOstoolsMcpServer). Tokens are injected server-side and the agent never holds them. A common egress-auth skill owns the wire contract (HTTP 511 auth_pending handshake, 403 deny), backoff polling, human-in-the-loop device-code relay, and a portable Python probe engine; thin user-invocable skills (connect, azure-devops, microsoft-graph) reference and invoke it with service-specific probe URLs, scopes, and endpoints.
Based on adoption, maintenance, documentation, and repository signals. Not a security audit or endorsement.
ALWAYS use this FIRST — before any Azure DevOps operation — to authenticate the sandbox. Trigger this before any request to dev.azure.com or *.visualstudio.com: listing/creating PRs, querying work items, reading repos/builds/pipelines, or any ADO REST call. The sandbox blocks unauthenticated egress, so ADO calls will fail until this runs. It triggers the egress proxy's auth handshake (may need a browser Azure AD login / consent, which it relays). Do NOT make Azure DevOps API calls until this reports SUCCESS.
ALWAYS use this FIRST — before calling ANY external service that does not have a dedicated skill — to authenticate the sandbox. Trigger this before the first request to a third-party API (Slack, Salesforce, OpenAI, Stripe, an internal service, etc.). The sandbox blocks unauthenticated egress, so calls will fail until this runs. Provide the service's probe URL (an endpoint that returns 2xx when authenticated). For GitHub use sandbox-auth:github, for Azure DevOps use sandbox-auth:azure-devops, and for Microsoft Graph use sandbox-auth:microsoft-graph. Do NOT call the target API until this reports SUCCESS.
Core sandbox egress authentication skill — the shared engine that the github / azure-devops / microsoft-graph / connect skills reference and invoke. Authentication is a REQUIRED FIRST STEP for any outbound call in the sandbox: the proxy blocks unauthenticated egress. Use whenever an outbound network call from inside a sandbox (git clone/fetch/push, gh, curl, wget, npm, pip, or a language HTTP client) targeting github.com, api.github.com, raw.githubusercontent.com, a package registry, or any external host returns an HTTP 511 (Network Authentication Required) with header `x-sandbox-proxy-status: auth_pending` or a body `{"status":"auth_pending",...}`, an HTTP 403 `{"error":"denied",...}`, or whenever a network command hangs, fails TLS, or behaves oddly behind the egress proxy. Also use PROACTIVELY before running token-unaware tools (git/gh/npm/pip) against GitHub to warm the auth cache. Owns the wire contract, the probe/poll engine, backoff polling, human-in-the-loop relay of login prompts, transparent server-side token injection, and the warm-then-run pattern. Other skills delegate the handshake mechanics here.
ALWAYS use this FIRST — before any GitHub operation — to authenticate the sandbox. Trigger this before: cloning/fetching/pushing a github.com repo (git), running `gh`, calling api.github.com / raw.githubusercontent.com via curl/wget/HTTP clients, or installing from GitHub with npm/pip. The sandbox blocks unauthenticated egress, so GitHub calls will fail until this runs. It triggers the egress proxy's auth handshake (may need an out-of-band / device-code login, which it relays) and warms the cache. Do NOT call GitHub or run git/gh until this reports SUCCESS.
ALWAYS use this FIRST — before any Microsoft Graph / Microsoft 365 operation — to authenticate the sandbox. Trigger this before any request to graph.microsoft.com: reading a user profile (/me), mail, calendar, OneDrive files, Teams, or any other M365/Graph data. The sandbox blocks unauthenticated egress, so Graph calls will fail until this runs. It triggers the egress proxy's auth handshake (may need a browser Azure AD login / consent or a ManualToken paste, which it relays). Do NOT make Graph API calls until this reports SUCCESS.
A curated marketplace of plugins for Claude Code, extending capabilities with custom skills, agents, commands, and tool integrations.
Add the marketplace to Claude Code:
/install-plugin https://github.com/gautam-achieveai/ClaudePlugins.git
Azure DevOps integration — work item management, PR publishing, iterative PR tending, autonomous work item implementation, and backlog processing.
Skills: ado-work-on, ado-publish-pr, ado-babysit-pr, ado-work-my-backlog, ado-draft-work-item, ado-pr-tender
GitHub integration — issue/project management, PR publishing, iterative PR tending, issue-driven implementation, and backlog processing.
Skills: gh-work-on, gh-publish-pr, gh-babysit-pr, gh-work-my-backlog, gh-pr-tender
Development methodology toolkit — design-first brainstorming, autonomous design (with review-planning and implementation-handoff steps), TDD, parallel subagent-driven execution with review gates, and evidence-based completion verification.
Skills: brainstorming, autonomous-design, test-driven-development, subagent-driven-development, verification-before-completion
Code review toolkit with specialized agents for duplicate detection, EUII leak scanning, exception handling review, test coverage review, design simplification, code simplification, over-engineering / scope-creep detection, architecture review, performance review, schema and wire-contract compatibility review (forward/backward compat, rollout sequencing, serializer asymmetry, DB migration footguns), feature-flag rollout review (blast-radius and reversibility), severity grading quality gate, log review, and PR work item context gathering. Includes review-pr command automation, structured PR comment publishing, and batch PR review orchestration with persistent tracking.
Skills: pr-review, post-pr-review, pr-context, review-pending-prs, update-pr-tracking, codebase-search-discipline, over-engineering-review, schema-compatibility-review
Evidence-based developer performance reviews over weeks/months. Analyzes git history, PRs, ADO work item context, bug patterns, and code quality. Auto-detects primary branch, two-dataset model (all work vs landed work), 6-dimension assessment framework.
Skills: developer-performance-review
Microsoft Orleans patterns, best practices, and code review for virtual actor model applications — grain design, concurrency, cross-grain communication, streams, and serialization.
Skills: orleans-patterns, orleans-code-review
Zero-warning builds through systematic warning elimination, code formatting (ReSharper, Roslynator, dotnet format), and NuGet package version validation.
Skills: clean-builds
Log-first debugging methodology using structured JSONL logs queried with DuckDB. Includes logging enablement for codebases, logging compliance review with calibrated Trace-coverage recommendations for AI-assisted debugging, and systematic root-cause debugging.
Skills: debug-with-logs, logging-enablement, systematic-debugging
claude-plugins/
├── .claude-plugin/
│ └── marketplace.json # Marketplace catalog (8 plugins)
├── ado/ # Azure DevOps integration
├── gh/ # GitHub integration
├── development/ # Dev methodology (brainstorming, TDD, etc.)
├── code-reviewer/ # PR code review toolkit
├── developer-performance-review/ # Developer performance reviews
├── orleans-dev/ # Orleans patterns & review
├── clean-builds/ # Zero-warning builds
├── debugging/ # Log-first debugging
├── scratchpad/ # Research and development notes
└── README.md # This file
Each plugin follows the standard structure:
plugin-name/
├── .claude-plugin/
│ └── plugin.json # Plugin manifest (name, version, description)
├── skills/
│ └── skill-name/
│ ├── SKILL.md # Skill entrypoint
│ ├── references/ # On-demand reference docs
│ └── scripts/ # Automation scripts
├── agents/ # Subagent definitions (optional)
├── commands/ # Slash commands (optional)
└── README.md
.claude-plugin/plugin.json with name, description, version.claude-plugin/marketplace.jsonThis marketplace is MIT licensed. Individual plugins may have their own licenses.
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 claimnpx claudepluginhub gautam-achieveai/claudeplugins --plugin sandbox-authWorkspace-agent skills for sandboxed task execution: bootstrap workspace memory, use scripts for repeatable work, maintain task state, and understand sandbox auth/context boundaries.
GitHub integration for issue/project management, PR publishing, iterative PR tending, issue-driven implementation, and backlog automation. Orchestrates development methodology skills via the development plugin.
Development methodology toolkit: design-first brainstorming, autonomous design with review-planning and implementation-handoff steps, test-driven development, parallel subagent-driven execution with review gates, and evidence-based completion verification. Includes planning, execution, and git workflow reference guides.
Azure DevOps integration for work item management, PR publishing, iterative PR tending, autonomous work item implementation, and autonomous backlog processing. Orchestrates development methodology skills via the development plugin.
Microsoft Orleans patterns, best practices, and code review for virtual actor model applications — covers grain design, concurrency, cross-grain communication, streams, and serialization.
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.
Harness-native ECC plugin for engineering teams - 67 agents, 271 skills, 92 legacy command shims, reusable hooks, rules, MCP conventions, and operator workflows for Claude Code plus adjacent agent harnesses
Develop, test, build, and deploy Godot 4.x games with Claude Code. Includes GdUnit4 testing, web/desktop exports, CI/CD pipelines, and deployment to Vercel/GitHub Pages/itch.io.
Core skills library for Claude Code: TDD, debugging, collaboration patterns, and proven techniques
Design fluency for frontend development. 1 skill with 23 commands (/impeccable polish, /impeccable audit, /impeccable critique, etc.) and curated anti-pattern detection.
UI/UX design intelligence. 84 styles, 161 palettes, 73 font pairings, 25 charts, 17 stacks (React, Next.js, Vue, Nuxt.js, Nuxt UI, Svelte, Astro, SwiftUI, React Native, Flutter, Tailwind, shadcn/ui, Jetpack Compose, Angular, Laravel, JavaFX, Three.js). Actions: plan, build, create, design, implement, review, fix, improve, optimize, enhance, refactor, check UI/UX code. Projects: website, landing page, dashboard, admin panel, e-commerce, SaaS, portfolio, blog, mobile app. Elements: button, modal, navbar, sidebar, card, table, form, chart. Styles: glassmorphism, claymorphism, minimalism, brutalism, neumorphism, bento grid, dark mode, responsive, skeuomorphism, flat design. Topics: color palette, accessibility, animation, layout, typography, font pairing, spacing, hover, shadow, gradient.