davila7/claude-code-templates

This system is a comprehensive CLI tool and web dashboard ecosystem designed to extend the capabilities of Claude Code through a modular architecture of agents, skills, hooks, and sandbox environments. It enables developers to automate workflows, enforce security and quality gates, and execute AI-generated code in isolated containers while providing a public-facing marketplace for discovering and installing these components. The platform also includes robust analytics, monitoring, and community management features to track usage and distribute updates.

31 changes.

This period focused on launching a comprehensive web-based analytics dashboard for the CLI tool, featuring real-time monitoring, modular backend architecture, and a 2025 Year-in-Review visualization. It also significantly expanded the platform's capabilities by introducing isolated execution sandboxes for Docker, Cloudflare Workers, and E2B, alongside a new Rust-based CLI core. The work was supported by extensive security validation systems, automated testing infrastructure, and a broad addition of specialized agent roles and productivity skills.

38 changes.

The project migrated its dashboard and infrastructure to Cloudflare, introducing a comprehensive web interface for browsing, curating, and installing components with features like collections and live terminal previews. Simultaneously, the CLI tool expanded significantly with new skills for AI research, security, web development, and specialized domains like DoorDash, alongside a new Rust-based installation core and a function hook system called Mods.

The dashboard now exposes new API routes under /api/collections to support creating, updating, and deleting user collections, as well as adding, removing, and moving items within them. Additionally, a new sharing feature allows users to toggle public visibility on a collection, generating a unique slug that enables unauthenticated access to view the collection's contents and author name via a dedicated public endpoint.

dashboard/src/pages/api/collections · high confidence

Added a new featured integration for Bright Data web data extraction, including a main skill file for the Bright Data MCP server (which replaces built-in WebFetch and WebSearch tools), a best-practices guide covering the Web Unlocker, SERP, Web Scraper, and Browser APIs, and a data-feeds skill for structured extraction from 40+ platforms like Amazon, LinkedIn, and Instagram. The entry also includes reference documentation for MCP setup (remote and local), tool catalogs, and detailed API usage patterns for authentication, request parameters, and error handling.

cli-tool/components/skills/web-data · high confidence

This change introduces a new Cloudflare-based sandbox environment for executing AI-generated code. It includes a Cloudflare Worker (\src/index.ts\) that integrates with the Anthropic API to generate and run Python or JavaScript code in isolated containers, along with a CLI launcher (\launcher.ts\) and a real-time monitoring tool (\monitor.ts\). The package also provides comprehensive setup and debugging documentation, configuration files for Wrangler and TypeScript, and environment variable templates to support local development and production deployment.

cli-tool/.claude, cli-tool/.claude/sandbox/cloudflare · high confidence

Users can now access AI-powered match predictions for the Premier League and Champions League through a new skill. This addition includes a shell script and documentation that allow fetching predictions for current or specific matchweeks, filtering by home or away teams, and viewing details such as half-time/full-time scores, next goal scorers, corner counts, and key player insights via the FootballBin MCP API.

cli-tool/components/skills/sports · high confidence

A new CLI skill for analyzing Google Analytics 4 data has been added to the analytics module. It includes Python scripts to connect to the GA4 Data API using service account credentials, fetch metrics and dimensions, and perform higher-level analysis such as period comparisons, traffic source breakdowns, content performance reviews, and conversion funnel tracking. The skill also provides configuration examples, usage documentation, and a metrics reference to help users interpret website performance and receive actionable improvement recommendations.

cli-tool/components/skills/analytics · high confidence

This change introduces the database schema, TypeScript types, and API endpoints required to support the new automated daily improvement cycle. It adds database tables to track review cycles, tool executions, and global pause controls, and exposes REST endpoints to manage cycle creation, status updates, tool logging, and administrative pause/resume functionality.

dashboard/src/lib/live-task, dashboard/src/pages/api/live-task · high confidence

This change introduces a new example todo application (backend and frontend) and Playwright utility helpers within the CLI tool's skills directory. The backend provides a REST API for managing todos using SQLite, while the frontend is a React application that consumes this API. Additionally, the Playwright skill now includes helper functions for browser automation, such as launching browsers, handling clicks and input, and taking screenshots.

cli-tool/components/skills/ai-research/loki-mode/examples/todo-app-generated/backend, cli-tool/components/skills/ai-research/loki-mode/examples/todo-app-generated/frontend, cli-tool/components/skills/utilities/playwright-skill · high confidence

The dashboard now includes static files for branding and device icons, specifically a new SVG logo depicting a terminal interface and a site.webmanifest defining Android Chrome icon sizes and theme colors.

dashboard/public/static · high confidence

The database now includes new tables and views to support two new monitoring capabilities. First, a changelog monitoring system tracks Claude Code versions, parsed changelog changes, and Discord notification logs to alert users of updates. Second, a command usage tracking system records raw CLI execution logs and automatically aggregates statistics, providing views for daily usage, platform distribution, and popular commands over the last 30 days.

database · high confidence

Added three new Git hooks that enforce workflow standards before commands execute: conventional commit validation ensures messages follow the type(scope): description pattern; branch name validation requires feature, release, or hotfix prefixes for new branches; and direct push prevention blocks pushes to main or develop branches, enforcing the use of pull requests.

cli-tool/components/hooks/git · high confidence

The dashboard is now available at app.aitmpl.com, providing a web interface to browse and install over 1,000 pre-built components for Claude Code. The site includes dedicated pages for skills, agents, commands, settings, hooks, MCP integrations, and autonomous loops, along with a new Jobs section for discovering relevant roles and a Component Review Loop for tracking automated improvements. A key addition is the Mods page, which offers early access to function-hook plugins (requiring Claude Code 2.1.259+ and a specific environment variable). The dashboard also features GitHub OAuth integration for user actions, dynamic SEO foundations including a sitemap and JSON-LD, and a dark-mode UI with a terminal-inspired aesthetic.

dashboard/src/pages · high confidence

The analytics web interface now includes a dedicated 2025 Year-in-Review view, presented as an interactive canvas-based activity graph with a heads-up display (HUD) for real-time statistics and a legend for tool/agent categorization. The main dashboard has been rebuilt using Astro, featuring a new bento-grid layout that displays token usage trends, session data, and agent analytics. Users can now filter data by time range (7D, 30D, 90D, All), toggle between light and dark themes, and view live updates via WebSocket connections. The dashboard also includes an activity heatmap and detailed breakdowns of input/output/cache token usage.

cli-tool/src/analytics-web · high confidence

Users can now execute AI-generated code in an isolated Cloudflare Workers environment. This new component includes a Cloudflare Worker (\src/index.ts\) that integrates with the Anthropic API to generate Python or JavaScript code and runs it in a sandboxed container, exposing \/execute\ and \/health\ endpoints. A CLI launcher (\launcher.ts\) and a real-time monitoring tool (\monitor.ts\) are provided for local development and debugging, alongside configuration files (\wrangler.toml\, \tsconfig.json\) and documentation to support deployment and usage.

cli-tool/components/sandbox/cloudflare · high confidence

The dashboard now includes a new Component Review Loop interface, allowing users to monitor automated daily improvement cycles. The LiveTaskPanel component provides a high-level view of cycle status (active, completed, merged, failed) and includes Open/All/Done filter pills to manage the list. The CycleTable component displays detailed information for each cycle, including the component path, type, current phase, associated PR, and duration. Additionally, the ToolTimeline component visualizes the sequence of tool executions (such as Read, Write, Edit, Bash, etc.) within a selected cycle, showing timestamps, tool names, and sanitized file paths to protect local directory structures.

dashboard/src/components/live-task · high confidence

Users can now execute Claude Code prompts in an isolated E2B cloud sandbox via the CLI, ensuring safe code generation without affecting the local system. This feature requires E2B and Anthropic API keys (provided via CLI arguments or environment variables) and supports installing additional components within the sandbox. The integration includes a launcher script with retry logic for WebSocket stability, extended timeouts (up to 15 minutes) to prevent premature termination, and automatic downloading of generated files to a local \e2b-output\ directory. A real-time monitor and debugging guide are also provided to help users track execution progress, system resources, and troubleshoot common issues like timeouts or permission errors.

cli-tool/components/sandbox/e2b · high confidence

The dashboard now includes a 'Stack Builder' feature, allowing users to curate a collection of components (agents, skills, hooks, etc.) and generate a single CLI command for installation. This is supported by a new CartSidebar component and an AuthButton that integrates with Clerk for user authentication and session management. The component grid has been redesigned to support filtering, sorting, and sponsored placements, while a new ComponentTerminal provides an animated, interactive preview of how each component installs and runs in the Claude Code CLI. Additional UI enhancements include a FileTreeSidebar for visualizing component file structures, a CountryFlag component for displaying user locations, and a FeaturedSection for highlighting integrations.

dashboard/src/components · high confidence

Users can now create, list, and remove global agents that are executable from any directory. The new global-agent-manager.js module handles downloading agent definitions from GitHub, saving them to a local templates directory, and generating executable scripts. It intelligently chooses between installing to the system binary directory (/usr/local/bin) for immediate availability or the user's local bin directory (requiring PATH configuration), ensuring agents are accessible via command line regardless of the current working directory.

cli-tool/src/sdk · high confidence

Added documentation and setup guides for several new AI research skills in the CLI tool. This includes agent evaluation and memory systems (agent-evaluation, agent-memory-mcp, agent-memory-systems), an agent manager for tmux-based session control, and an agent tool builder for designing LLM tool schemas. It also introduces comprehensive guides for major agent orchestration platforms: AutoGPT (including advanced block development and troubleshooting), CrewAI (covering flows, tools, and troubleshooting), and LangChain (covering agents, integrations, and vector stores).

cli-tool/components/skills/ai-research · high confidence

The dashboard now exposes several new API routes to support internal monitoring and user analytics. The \/api/claude-code-check\ endpoint monitors the official Claude Code npm package for new releases, parses the changelog, and sends formatted notifications to a configured Discord webhook. A new \/api/health-check\ route periodically verifies the availability of external tracking endpoints and alerts the team via Discord if any failures occur. Additionally, new tracking endpoints have been added to capture specific user interactions: \/api/track-command-usage\ logs CLI command executions, \/api/track-installation-outcome\ records the success or failure of component installations, \/api/track-download-supabase\ tracks component downloads with geographic and device data, and \/api/track-website-events\ ingests batched website interaction events. All endpoints are implemented as Astro API routes and store data in the database or Supabase.

dashboard/src/pages/api · high confidence

The \create-claude-config\ command is introduced as the primary CLI entry point, replacing previous hardcoded versioning with dynamic detection from package.json. It provides a comprehensive set of flags for managing Claude Code configurations, including template selection, directory targeting, and component installation (agents, commands, hooks, skills, mods). The command also exposes direct access to various dashboards and interfaces such as analytics, chats, agents, plugins, skills, teams, and the Studio interface, alongside utilities for health checks, session sharing, and sandbox execution.

cli-tool/bin · high confidence

The project has migrated its automation and monitoring infrastructure from Vercel to Cloudflare Workers, introducing several new edge-based capabilities. A new \crons\ worker replaces previous Vercel cron jobs, scheduling health checks and release monitors with Sentry error tracking. A \daily-health-report\ worker provides a daily Telegram digest of site health and Sentry issues, featuring conservative auto-resolution of known test noise. A \docs-monitor\ worker tracks changes to the Claude Code documentation and notifies via Telegram. A \pulse\ worker generates weekly KPI reports using metrics from GitHub, Discord, and npm. Finally, a \newsletter\ worker sends a weekly community email via Resend, featuring weighted-random selection of trending components and rotating copy.

cloudflare-workers · high confidence

The document-processing area now includes two new skills: a generic 'doc' skill for reading, creating, and editing .docx files with visual layout checks (using python-docx and LibreOffice/PDF rendering), and an 'official' docx skill providing comprehensive creation, editing, and analysis capabilities (including tracked changes, comments, and styling) via docx-js and Python scripts. Additionally, a 'documentation-templates' skill has been added to provide structured templates for READMEs, API docs, code comments, changelogs, and architecture decision records.

cli-tool/components/skills/document-processing · high confidence

Users can now run Claude Code inside an isolated Docker container using the new \--sandbox docker\ flag. This feature introduces a launcher script that builds and manages a Node.js 22 Alpine image containing the \@anthropic-ai/claude-agent-sdk\, executing prompts with auto-approved permissions and saving generated files to a local output directory.

cli-tool/components/sandbox/docker · high confidence

This location introduces a suite of five new skills for the DoorDash CLI (dd-cli) that add safety, coordination, and accountability to agent-driven food ordering. The doordash-allergy-shield skill persists a dietary profile and enforces a mandatory cart-vetting step with a deterministic hook that blocks checkouts containing high-severity allergens. The doordash-group-orders skill manages team rosters, attributes cart items to specific people, and calculates cost splits for shared orders. The doordash-spend-guard skill enforces hard budget limits (per-order, daily, weekly, monthly caps, cooldowns, and allowed hours) via a wrapper script (dd-guard.sh) that intercepts cart mutations and checkouts, refusing out-of-policy actions. The doordash-order-playbooks skill allows saving and recalling named orders while detecting menu or price drift before checkout. Finally, the doordash-order-ledger skill provides an append-only audit log of all dd-cli commands and a reporting command to answer spending and activity questions from data rather than memory.

cli-tool/components/skills/doordash · high confidence

This update adds three new skills to the Git toolkit. The 'commit-smart' skill analyzes staged and unstaged changes to automatically generate semantic conventional commit messages with context. The 'git-context-controller' skill introduces a versioned file system under .GCC/ to persist agent memory, allowing users to commit milestones, branch for alternative approaches, merge results, and recover cross-session context. The 'star-history-chart' skill provides a self-hosted solution for displaying stargazer history in READMEs by generating a static, theme-aware SVG via a Python script and a weekly GitHub Action, bypassing third-party service authentication restrictions.

cli-tool/components/skills/git · high confidence

The CLI tool now supports a new 'Mods' system (function hooks) that allows plugins to hook into the engine's events, requiring Claude Code \>= 2.1.259 and the \CLAUDE\_CODE\_ENABLE\_FUNCTION\_HOOKS=1\ environment variable. This location introduces the \mods\ directory structure, documentation, and several bundled mods: an enterprise \admin-capability-lockdown\ mod that restricts network/process access and plugin registration, and game mods including \2048\ and the full \cc-arcade\ suite (snake, Tetris, Doom, etc.) that run above the prompt.

cli-tool/components/mods · high confidence

A new 'Comprehensive OWASP Security Skill' has been added to the .claude-plugin/skills directory. This skill provides developers with unified security guidance across six OWASP standards: Top 10 (2021), ASVS 5.0, MASVS v2.1.0, API Security Top 10 (2023), Kubernetes Top 10 (2022), and Agentic Applications 2026. It includes a detailed reference guide (SKILL.md) and nine intentionally vulnerable code examples (covering web, API, mobile, Kubernetes, and AI/LLM contexts) that demonstrate insecure patterns alongside secure implementations to help users identify and remediate security risks in their code, configurations, and architecture.

.claude-plugin · high confidence

Added a suite of Python scripts to the Obsidian Ops Team agent to automate vault maintenance. The new tools include a Link Suggester and entity/keyword connection implementers to automatically create bidirectional links between notes based on shared topics and entities; a Map of Content (MOC) generator to create structured overview pages for directories; a Daily Notes Connector to link daily entries by date and content; an Enhanced Tag Standardizer to consolidate and hierarchically organize tags; a Metadata Adder to inject standardized frontmatter into files lacking it; and utility scripts to fix quoted tags and parse connection reports.

cli-tool/components/agents/obsidian-ops-team · high confidence

The dashboard now includes dedicated pages for browsing and installing Claude Code plugins and marketplaces. Users can visit the main Plugins page to view a grid of curated collections and use the integrated search modal, or navigate to individual plugin detail pages to view installation commands, metadata, and links to source repositories. These pages are statically generated from the \public/plugins.json\ data file.

dashboard/src/pages/plugins · high confidence

The CLI tool now includes a new web-based analytics dashboard built with Astro and React. This UI provides a visual interface for monitoring Claude Code usage, featuring real-time data updates via WebSocket, interactive charts (token usage timelines, activity heatmaps, agent distributions), and a mobile-optimized chat view. The dashboard includes a dark/light theme toggle, time-range filtering (7d, 30d, 90d, all), and detailed session tracking. It is served as a static site from the CLI's analytics-web directory.

cli-tool/src · high confidence

This change introduces a suite of new scripts to the \scripts/\ directory to support the project's operational and content needs. A new \deploy.sh\ script automates the deployment of the Astro dashboard (serving both www.aitmpl.com and app.aitmpl.com) to Vercel using environment variables. Several Python scripts are added to generate static content for the dashboard: \generate\_agents\_api.py\ creates an agents API endpoint, \generate\_plugins\_json.py\ fetches and aggregates plugin data from various GitHub repositories, \generate\_claude\_jobs.py\ scrapes job listings, \generate\_claude\_prs.py\ tracks Claude Code pull requests, and \generate\_star\_history.py\ produces a self-hosted stargazer chart. Additionally, \generate\_blog\_images.py\ and \generate\_blog\_images\_v2.py\ automate the creation of blog cover images using Google's Imagen API, and \generate\_components\_json.py\ handles component data generation including security validation. A \dev-server.js\ script is also added to serve the static documentation files locally for development.

scripts · high confidence

This update adds a suite of new hook components to the automation toolkit. Users can now automatically load AGENTS.md configuration at session start, trigger builds on file changes, and monitor deployment health via Vercel. The toolkit introduces comprehensive logging of file mutations to CSV and adds notification integrations for Discord, Slack, and Telegram, offering both simple finish alerts and detailed session summaries (duration, memory, project info). Additionally, a playful 'Plan Mode Game' feature launches browser games during planning phases, and a dependency checker runs security audits when package files are modified.

cli-tool/components/hooks/automation · high confidence

The CLI tool now includes new skills in the business-marketing category to guide users through A/B testing setup, analytics tracking implementation, and AI-powered product architecture. The ab-test-setup skill provides frameworks for hypothesis generation, sample size calculation, and metric selection for split tests. The analytics-tracking skill offers event naming conventions and essential event definitions for marketing sites, products, and e-commerce. The ai-product and ai-wrapper-product skills cover LLM integration patterns, prompt engineering, cost management, and output validation for building production-ready AI applications. Additionally, the agile-product-owner skill adds user story generation capabilities with INVEST criteria validation, and the app-builder orchestrator now includes 13 project templates covering Next.js, Nuxt, Express, FastAPI, React Native, Flutter, Electron, Chrome extensions, CLI tools, and monorepo setups.

cli-tool/components/skills/business-marketing · high confidence

A new 'career' skills category has been added to the CLI tool, providing a suite of specialized capabilities for the job-search workflow. This update introduces distinct skills for analyzing job descriptions and calculating match scores, generating interview preparation materials like STAR stories, optimizing LinkedIn profiles for recruiter visibility, and creating tailored application documents including academic CVs, executive resumes, cover letters, and creative portfolio resumes. It also includes tools for translating skills across industries for career changers, comparing multiple job offers with total compensation analysis, and writing detailed portfolio case studies.

cli-tool/components/skills/career · high confidence

The creative-design skill set has been expanded with new capabilities: the 3D Web Experience skill provides guidance for building interactive 3D scenes using Three.js, React Three Fiber, and Spline; the Accessibility Auditor skill offers detailed patterns for WCAG 2.1 compliance, ARIA implementation, and semantic HTML; the Algorithmic Art skill enables the creation of generative p5.js art with seeded randomness and interactive parameter controls; and the C4 Architecture skill generates software architecture documentation using Mermaid C4 diagrams across context, container, component, and deployment levels.

cli-tool/components/skills/creative-design · high confidence

The dashboard now includes a new set of API helper modules in \src/lib/api\. Authentication is handled via Clerk, with functions to verify tokens and retrieve user emails. A changelog parser has been added to parse version-specific changes, classify them (features, fixes, improvements, etc.), and format summaries for platforms like Discord. Error tracking is implemented using a lightweight Sentry reporter that sends envelopes directly without an SDK dependency. Additionally, CORS headers and JSON response helpers are standardized, and a Neon database client factory is provided for serverless database access.

dashboard/src/lib/api · high confidence

The dashboard now includes a new set of library modules in \src/lib\ that power several new capabilities. \collections-api.ts\ provides the client-side API for creating, renaming, and managing user collections of components. \ads.ts\ introduces a feature-flagged (PUBLIC\_ADS\_ENABLED) system for fetching and displaying sponsored placements from the \/api/ads/active\ endpoint, ensuring search and grids remain functional even if the ad service fails. \webmcp.ts\ exposes the component catalog as read-only tools for browser-based AI agents via the WebMCP standard, allowing agents to search and retrieve component details. Additionally, \data.ts\ implements a split-data loading strategy with caching to reduce browser payload size, \constants.ts\ defines featured integrations and navigation links, \icons.ts\ provides type-specific SVG icons, \theme.ts\ manages dark/light mode persistence, and \github-api.ts\ enables direct GitHub repository listing and PR creation.

dashboard/src/lib · high confidence

A new 'design-to-code' skill has been added to the CLI tool, enabling the conversion of Figma designs into production-ready React, TypeScript, and TailwindCSS components. This feature introduces a helper script (\coderio-skill.mjs\) and a structured workflow (Setup, Protocol Generation, Code Generation) that automates fetching Figma data, generating component structures and props, and producing the final code. Users can now leverage this automated pipeline to achieve high-fidelity UI restoration from design files.

cli-tool/components/skills/design-to-code · high confidence

Added two new skills to the development toolkit: an accessibility skill that provides WCAG 2.1 guidelines and audit procedures for improving web accessibility, and an agent-development skill that includes templates, system prompt design patterns, and validation scripts for creating autonomous agent plugins.

cli-tool/components/skills/development · high confidence

The CLI tool now includes a set of new specialized agent definitions within the development-team directory, expanding the available AI personas for software engineering tasks. New roles include backend-architect for system design and API paradigm selection, backend-developer for implementation of server-side services, code-architect for feature blueprints, code-explorer for codebase analysis, devops-engineer for CI/CD and infrastructure, electron-pro for desktop application development, flutter-ui-developer for cross-platform mobile/web UIs, and frontend-developer for modern web frameworks (React, Vue, Angular). Additionally, a new cli-ui-designer agent has been added to create terminal-inspired user interfaces using web technologies. These agents provide structured workflows, specific tooling recommendations, and best-practice guidelines for their respective domains.

cli-tool/components/agents/development-team · high confidence

Added a suite of new hook configurations and scripts in the development-tools directory to enhance the development workflow. The debug-window hook now automatically opens a live-tailing debug log viewer on session start (when the --debug flag is used) and closes it on session end, with an option to disable auto-closing. New audit and tracking capabilities include change-tracker for logging file modifications, command-logger for recording all tool usage, and edit-audit-log for tracking edits to a project-local file. Workflow automation is expanded with file-backup to create timestamped backups before edits, smart-formatting to auto-format code using language-specific tools (Prettier, Black, gofmt, etc.), and lint-on-save to run linters like ESLint and Pylint after changes. Additionally, a Next.js code quality enforcer validates best practices for App Router components, and a worktree-ghostty hook automates the creation of git worktrees and opens a 3-panel Ghostty terminal layout with lazygit and yazi on macOS.

cli-tool/components/hooks/development-tools · high confidence

The dashboard now includes a new dynamic page for individual featured integrations (located at \dashboard/src/pages/featured/\[slug\].astro\). This page renders a detailed view for each integration using data from \FEATURED\_ITEMS\, featuring a redesigned layout with a verified badge, social proof indicators, and a prominent call-to-action. It also includes a sticky navigation bar and a premium hero banner with animated backgrounds, providing a more polished and informative experience for users exploring specific integrations like Bright Data or ClaudeKit.

dashboard/src/pages/featured · high confidence

Added new skills for backend-to-frontend API handoff documentation, daily meeting standup generation with GitHub/Jira integration, CAPA (Corrective and Preventive Action) management, data privacy compliance (GDPR, CCPA, HIPAA), and Anthropic brand styling guidelines.

cli-tool/components/skills/enterprise-communication · high confidence

The CLI tool now includes three new media skills in the \cli-tool/components/skills/media\ directory. The \image-enhancer\ skill improves image quality by upscaling resolution, sharpening edges, and reducing compression artifacts. The \screenshot\ skill provides cross-platform desktop capture capabilities (macOS, Linux, Windows) with support for full-screen, window-specific, and region-based captures, including a macOS preflight script to handle Screen Recording permissions. The \speech\ skill enables text-to-speech generation using OpenAI's \gpt-4o-mini-tts-2025-12-15\ model by default, with optional support for Atlas Cloud TTS as an alternative backend.

cli-tool/components/skills/media · high confidence

Added three new hook configurations to the monitoring directory: a context-timeline visualizer that launches a local HTTP server to display real-time context window usage and tool execution as a git-graph timeline; a desktop notification hook that sends native OS alerts (via osascript or notify-send) when a session stops; and a LangSmith tracing integration that exports conversation traces to the LangSmith API for monitoring. These hooks enable users to visualize session activity, receive completion alerts, and integrate with external observability tools.

cli-tool/components/hooks/monitoring · high confidence

This update adds several new productivity skills to the CLI tool, including the Academy Guide (which recommends learning resources from Claude Academy), Avoid AI Writing (which audits and rewrites content to remove AI writing patterns), Address GitHub Comments (for handling PR feedback via the gh CLI), Brainstorming (for collaborative design exploration), Claude Code Sessions (for searching and managing session history), Code Review Excellence (for structured code review feedback), Code Simplifier (for refactoring code for clarity), and Commit Work (for creating high-quality git commits).

cli-tool/components/skills/productivity · high confidence

The CLI tool now includes dedicated skill sets for three programmatic video creation frameworks. Users can generate mathematical and educational animations using Manim (Python), create motion graphics with TypeScript via Motion Canvas (including specific setup guidance for ESM/CommonJS interoperability), and build video compositions using React with Remotion. These skills provide comprehensive reference documentation, best practices, and code examples for scenes, animations, assets, and framework-specific constraints.

cli-tool/components/skills/video · high confidence

Added three new pre-commit hooks in the quality-gates directory to support Spec-Driven Development and Test-Driven Development workflows. The plan-gate hook warns when editing production code without a recent specification file (.spec.md). The scope-guard hook detects and warns about files modified outside the declared scope of an active specification. The tdd-gate hook enforces Test-Driven Development by blocking edits to production code files unless a corresponding test file already exists. All hooks are non-blocking except tdd-gate which blocks when no test is found, and they support multiple programming languages including C\#, TypeScript, JavaScript, Python, Go, Rust, Ruby, PHP, Java, Kotlin, Swift, and Dart.

cli-tool/components/hooks/quality-gates · high confidence

Added new skill definitions for the Adaptyv cloud laboratory platform (enabling protein testing, validation, and sequence optimization workflows) and the Aeon time series machine learning toolkit (covering classification, regression, clustering, forecasting, anomaly detection, and similarity search). These additions expand the CLI tool's scientific capabilities with detailed usage instructions, API references, and code examples for both domains.

cli-tool/components/skills/scientific · high confidence

Added new security skills to the CLI tool for conducting penetration tests and security assessments on Active Directory, AWS, Azure, and general cloud environments, as well as for API security testing (including fuzzing, bug bounty techniques, and best practices) and broken authentication analysis. These skills provide structured workflows, tool recommendations, and exploitation techniques for authorized security engagements.

cli-tool/components/skills/security · high confidence

Added a suite of new security hooks in the CLI tool to prevent accidental data loss and credential exposure. The 'dangerous-command-blocker' and 'shell-wrapper-guard' hooks intercept Bash commands to block catastrophic operations (like recursive deletes or disk formatting) and detect destructive patterns hidden inside interpreter wrappers (e.g., python3 -c, sh -c). The 'secret-scanner' hook prevents commits containing hardcoded API keys, tokens, and passwords from major providers. Additionally, 'force-push-blocker' stops accidental git history overwrites, while 'file-protection' and 'env-file-protection' hooks prevent the modification or overwriting of critical system files and environment configuration files.

cli-tool/components/hooks/security · high confidence

The CLI tool now includes a comprehensive security validation system for agents, commands, MCPs, settings, and hooks. This new feature introduces a five-tier validation framework: Structural (YAML frontmatter, file size, encoding), Semantic (prompt injection, jailbreaks, credential harvesting), Reference (URL protocol validation, private IP blocking), Integrity (SHA256 hash generation and tamper detection), and Provenance (author metadata, Git commit tracking). The system is orchestrated by ValidationOrchestrator and accessible via the --security-audit CLI flag, which can run in CI/CD mode, generate verbose output, or produce JSON reports. This adds a new layer of security assurance to component distribution.

cli-tool/src/validation · high confidence

Added a large set of new statusline configurations to the CLI tool, including specialized monitors for game development workflows (asset pipeline, Unity/Unreal/Godot performance, multi-platform builds), Neon database context and resource tracking, real-time Vercel and Cloudflare Pages deployment monitoring with clickable links, and a context usage monitor with visual progress bars. Also included several playful, gamified statuslines (casino, fitness tracker, RPG, tamagotchi, circus) and standard utility lines (minimal, git branch, deadline countdown, project info).

cli-tool/components/settings/statusline · high confidence

The CLI tool now includes several new skills in the utilities directory to assist with development and deployment tasks. Users can now automate browser interactions using Playwright, which auto-detects local dev servers and executes test scripts. A new Cloudflare Crawl skill allows users to ingest website content into markdown format using the Browser Rendering API, with support for incremental crawls via the \modifiedSince\ parameter and automatic credential loading from \.env\ files. Additionally, a Domain Name Brainstormer skill helps generate and check availability for domain names across multiple TLDs, and a Browser Extension Builder skill provides guidance on creating Manifest V3 extensions. Other new utilities include a BusyBox-on-Windows helper, a Network 101 skill for setting up penetration testing labs, and a GEO Fundamentals skill with a Python script to audit web pages for AI citation readiness.

cli-tool/components/skills/utilities · high confidence

The CLI tool now includes a set of new expert skills in the web-development category to guide users through specific modern frameworks and tools. These additions cover Astro for content-focused sites with islands architecture, BlockRun for external model capabilities like image generation and real-time X data, and a comprehensive workflow for building blogs with Next.js and Sanity CMS (including technical requirements and image style guides). Additional skills provide expertise in Chrome Extension development (Manifest V3), Drizzle ORM for type-safe database layers, Electron for secure desktop apps, and Exa for semantic search. These skills are designed to trigger automatically when users ask about these specific technologies, providing structured, best-practice guidance.

cli-tool/components/skills/web-development · high confidence

A new public-facing page at /c/{slug} allows users to view and install components from shareable collections. The page fetches collection metadata and items server-side, displays the collection title and author, groups components by type (e.g., agents, skills, commands), and provides a one-click install command for all components in the collection.

dashboard/src/pages/c · high confidence

The \cli-rust\ directory introduces a Rust implementation of the core component installation logic (agents, commands, MCPs, settings, hooks, and skills), ensuring byte-for-byte parity with the existing Node.js CLI. This new binary, distributed as \cct\, handles the native install path while delegating all other commands (such as dashboards, sandbox, and global agent management) back to the Node CLI via \npx\ or a configurable \CCT\_NODE\_BIN\ environment variable. The package also includes an npm wrapper (\@davila7/cct\) that acts as a thin launcher shim to execute the prebuilt Rust binary for the current platform.

cli-rust · high confidence

A new hook script has been added to automatically send a Telegram notification whenever a pull request is created. The notification includes the PR URL, the current branch name, and, if configured, the Vercel preview deployment URL, allowing teams to stay informed about new PRs directly in their chat.

.claude/hooks · high confidence

The analytics backend has been restructured from a monolithic file into a modular architecture, introducing dedicated classes for conversation loading (ConversationAnalyzer), file watching (FileWatcher), process detection (ProcessDetector), state calculation (StateCalculator), and session analysis (SessionAnalyzer). This refactor includes a new AgentAnalyzer to track specialized agent usage patterns and a YearInReview2025 module for yearly statistics. A multi-level DataCache system has been added to optimize performance and reduce memory usage by caching file content, parsed conversations, and computed results. Additionally, the system now detects both 'Task' and 'Agent' tool names for analytics and provides real-time updates via a NotificationManager.

cli-tool/src/analytics · high confidence

The blog for www.aitmpl.com is now served directly from the dashboard's public directory, replacing the previous external hosting. This change introduces a JSON-based content management system where articles are defined in \blog-articles.json\ and rendered dynamically via \blog-loader.js\. The update includes a comprehensive README detailing how to add and manage articles, along with a collection of new SVG cover images for topics such as Function Hooks, security workflows, and various agent skills.

dashboard/public/blog · high confidence

The dashboard has been visually redesigned with a terminal-inspired aesthetic, introducing a dark-mode-first color palette and monospace typography across the interface. This update adds dedicated styling for several new content areas: a blog section with article cards, a component library page with back-navigation and loading states, a featured page with partner badges, and a partnership banner. It also introduces detailed layouts for plugin and stack detail pages, a trending analytics view with growth charts, and a workflow builder with a component tree and modal previews.

dashboard/public/css · high confidence

The dashboard now serves its static data files (components, counts, search index, plugins, and trending data) with explicit 24-hour edge caching and cross-origin access headers, while Astro assets are marked immutable for long-term caching. To support the renaming of 'Function Hooks' to 'Mods', old URLs are permanently redirected (301) to the new paths. Additionally, strict security headers (HSTS, X-Frame-Options, Referrer-Policy, Permissions-Policy) are applied to all routes to harden the dashboard's security posture.

dashboard/public · high confidence

The dashboard has been rebuilt using Astro with React and Tailwind CSS and migrated from Vercel to Cloudflare Pages. This change introduces support for Clerk authentication, GitHub OAuth for the 'Send to Repo' feature, and Discord bot interactions. It also enables sponsored ad placements (controlled by the PUBLIC\_ADS\_ENABLED flag) and WebMCP catalog tools (controlled by PUBLIC\_WEBMCP\_ENABLED), while configuring the environment for Cloudflare's runtime via wrangler.toml and middleware.

dashboard · high confidence

New pre- and post-execution hooks have been added to the DoorDash CLI integration to enforce safety and provide transparency. The allergy-checkout-gate and group-checkout-gate hooks now block checkouts if anaphylaxis-tier allergen matches are detected in the cart or group roster, ensuring deterministic protection against severe allergic reactions. A spend-guard hook prevents budget bypasses by requiring all checkout and cart-addition commands to go through the dd-guard.sh wrapper, which enforces spending policies. Additionally, an audit-log hook passively records every CLI invocation to an append-only JSONL file, allowing users to review command history and costs without altering session behavior.

cli-tool/components/hooks/doordash · high confidence

The dashboard now uses a new \DashboardLayout\ component that establishes the core page structure, including a collapsible sidebar with persistent state, a mobile-responsive overlay, and a footer acknowledging open-source sponsors. This layout integrates critical SEO foundations by injecting meta tags, canonical links, Open Graph/Twitter cards, and JSON-LD structured data. It also supports a dark mode UI overhaul via CSS variables and a theme initialization script, while preloading the LCP image and initializing WebMCP agent tools.

dashboard/src/layouts · high confidence

The component detail pages now use a unified layout that dynamically selects the appropriate viewer (Markdown or JSON) based on the component type, significantly improving the presentation of settings, hooks, and MCPs. The page includes a new live Claude Code terminal in the hero section to demonstrate component usage, displays metadata cards for key attributes, and supports Loops by parsing and linking to referenced components. Additionally, the page structure has been optimized to fetch content on demand, reducing the initial payload size by approximately 87%.

dashboard/src/pages/component · high confidence

Added comprehensive test coverage for analytics, validation, and configuration components; New testing infrastructure and publishing configuration for the CLI tool.

This change adds a new Rust-based CLI core (\cli-rust\) with its \Cargo.toml\ and \Cargo.lock\, establishing a native binary distribution alongside the existing Node.js tooling. It also introduces dependency manifests for several new sandbox providers (Cloudflare Workers, Docker, and E2B) and various skill scripts (including Playwright, Lighthouse, and Google Analytics), while updating the main \cli-tool\ package to version 1.29.4 and its analytics UI to use Astro 5 and React 19.

(dependencies) · high confidence

Written by watchdog.canine.dev from the codebase's own history, inside the signed delivery this page is composed from.

• First survey — no prior run to compare against. CAI 48.

• Code Health 41
• Architecture 87
• Maturity 84
• Readiness 45
• Security 67
• Accessibility 51

• 300 commits — 288 feature/other, 12 fixes

• dashboard/public — 157 commits
• cli-tool/components — 107 commits
• (root) — 9 commits
• dashboard/src — 9 commits
• docs/star-history.svg — 9 commits
• docs/blog — 3 commits
• scripts/generate_plugins_json.py — 2 commits
• .github/workflows — 1 commit
• api/claude-code-monitor — 1 commit
• cli-tool/docs_to_claude — 1 commit
• cloudflare-workers/newsletter — 1 commit

• fix: Rename risk-manager to trading-risk-manager, fix naming collision (#765)
• fix: fix(blog): correct Function Hooks article against a real 2.1.266 binary (#875)
• fix: fix(dashboard): show Featured Integrations only on the home page (#872)
• fix: fix(mcps): refresh trace-mcp counts and qualify the token-reduction claim (#844)
• fix: fix(plugins): register krasserm/ml-plugins in generator REPOS, restoring ml-research listing (#613, dropped by #697) (#718)
• fix: fix(readme): switch stargazers chart to star-history.com
• fix: fix: add missing 'item' field to last breadcrumb ListItem for Search Console (#732)
• fix: fix: narrow discernment-nudge code-execution exemption (#900)
• fix: fix: repair invalid YAML frontmatter in six skills (#818)
• fix: improve: enhance ai-ethics-advisor agent (fix frontmatter, add current standards) (#772)
• fix: improve: enhance shopify-expert agent (2026 currency + frontmatter fix) (#691)
• fix: improve: fix droid agent to use Claude Code conventions instead of GitHub Copilot (#781)
• change: Add Cohesivity plugin marketplace listing (#827)
• change: Add Curviate skills and MCP server (#913)
• change: Add Dim Hour MCP to web-data (#830)
• change: Add HumanPen MCP under productivity (#796)
• change: Add Live Tennis API MCP (integration) (#813)
• change: Add OrcaReplay MCP component (devtools) (#863)
• change: Add avoid-ai-writing skill (productivity) (#773)
• change: Add blog post: Performance Budget Guard hook (#854)
• …and 280 more

Written by watchdog.canine.dev from the codebase's own history, inside the signed delivery this page is composed from.

davila7/claude-code-templates was measured the same way every project in this corpus was: the same rubric, at a pinned commit, with the result published in full. Point a surveyor at a repository you know and see whether you agree with it.