Discover +336 AI Coding apps & tools

Pros: Real-time hit/miss analytics reveal cache behavior per session. Automated cache_control breakpoint injection reduces manual cache logic. Native MCP integration plugs into Claude Desktop and Cursor. Open-source codebase enables inspection and community contributions.

Cons: Limited to Anthropic models that support prompt caching. Requires an MCP-capable client plus a valid Anthropic API key. Session-level savings reporting may not reflect organization-wide usage.

Pros: Programmatic access to project internals for automated audits. Supports live editor routes and headless manipulation via MCP. Read-only HTTP dashboard provides real-time project status. Designed specifically for Godot 4.x projects and workflows.

Cons: Requires an MCP-compatible client to connect. Limited to Godot 4.x, not backward compatible with Godot 3.x. Server process setup adds deployment overhead for small teams. Generated edits require manual verification before committing.

Pros: Supports OpenAI, Anthropic, Google Gemini, and Mistral APIs. Native Ollama support enables local inference and offline runs. Acts as a Model Context Protocol server for editor integrations. Configurable via CLI commands or environment variables.

Cons: Requires a Node.js environment and npm or yarn familiarity. Command-line interface assumes developer experience, not casual users. Plugin extensibility requires custom development to add tools.

Pros: Native Swift implementation using macOS system APIs. Exposes shortcuts as standard MCP tools for compatible clients. Runs locally, keeping shortcut data and execution on the host. Open-source codebase permits inspection and community contributions.

Cons: Requires macOS 14.5 or later to operate. Building from source requires Xcode 16.x. AI can trigger shortcuts but cannot inspect their internal logic. Only works with AI clients that support the Model Context Protocol.

Pros: OpenAI-compatible single endpoint eases migration for existing client libraries. Built in Rust, low resource overhead during API routing. Health monitoring exposes server availability for operational visibility. Supports local, air-gapped deployments to keep data on-premises.

Cons: Depends on LlamaEdge-compatible API servers, requiring backend setup. Custom web UI requires configuration before it is served. Gateway preserves backend outputs, necessitating independent model validation.

Pros: Exposes Frida functions to AI clients via the Model Context Protocol. Automated TypeScript agent scaffolding reduces boilerplate for new projects. Integrated REPL enables immediate script testing and iterative debugging. Unified CLI centralizes server, script, and process control.

Cons: Requires local Python 3.x and Node.js toolchain. Generated hooks need manual validation on complex targets. Functionality depends on correct Frida server setup on targets.

Pros: Processes data locally inside the browser, avoiding remote uploads. Single-file distribution supports offline use and fast loading. MCP integration lets AI models call specific utilities during sessions.

Cons: Large catalog requires time to locate the right utility. Not intended as production cryptographic key management. Interface density can overwhelm newcomers without curation.

Pros: Hot reloading applies saved script changes without restarting the server. Sandboxed Starlark runtime enforces deterministic, isolated execution. Built-in modules for HTTP, SQL, JSON, and time simplify integrations. Single portable executable across major desktop platforms.

Cons: Requires an MCP-compliant client for full functionality. System command execution requires explicit whitelist configuration. Starlark's simplified dialect omits some Python standard behaviors.

Pros: Local MCP server exposes saved snippets to desktop AI assistants. Supports JavaScript, Python, and Rust snippet storage. Native desktop client with automatic light and dark theme following. One-click clipboard integration for fast insertion into editors.

Cons: Requires a compatible desktop AI client to unlock AI-context features. Benefit depends on the quality and configuration of the external assistant. No cloud sync described, limiting seamless multi-device access.

Pros: Maintains a live shell session so state persists across turns. Provides structured table outputs that aid model parsing. Built-in modules for Kubernetes, Tmux, and Git extend automation. Rust implementation improves performance and memory safety.

Cons: Requires Nushell installed and present on the system PATH. Sandboxing reduces risk but does not replace manual command review. Needs an MCP-compliant client to connect, such as a desktop client.

Pros: Exposes GNS3 through MCP for direct LLM interaction. Supports CRUD plus batch and wildcard node operations. SSH automation for over 200 device types, multi-vendor coverage.

Cons: Needs a running GNS3 instance to operate. Automated outputs require operator validation before deployment. Optimized for Windows; other hosts need Docker or Python MCP setup.

Pros: Causal chain analysis links CPU events to GPU execution.. Sub-microsecond tracing captures kernel-to-CUDA timelines.. Runs without modifying application code or container images.. MCP server lets AI agents query performance data directly..

Cons: Limited to NVIDIA GPUs in the CUDA ecosystem.. Requires Linux hosts for deployment.. Agent access to traces requires deliberate access controls.. Automated recommendations need human validation before rollout..

Pros: Isolates workspaces using a git worktree-first architecture. Persistent session saving restores full conversation context. Includes a protocol trace viewer for MCP debugging. Live token usage tracking displays API consumption.

Cons: Windows requires WSL rather than native support. Not a replacement for Claude Code; it augments the CLI. Effective use requires familiarity with git worktrees and CLI workflows.

Pros: MCP-native toolset tailored for LLM-style context requests. Docker and native Linux deployment options for flexible hosting. Lists and retrieves Polarion custom fields and revision histories. Uses HTTPS with API-key authentication for secure transport.

Cons: Requires Polarion instance URL and a valid Personal Access Token. Output fidelity depends on the Polarion REST API responses. Initial setup needs config.json and deployment planning.

Pros: Processes and indexes code locally, avoiding external uploads. Presents a senior-engineer perspective on project structure to models. Fast query resolution on large repositories with a low resource footprint. Integrates with MCP-capable clients and CLI workflows.

Cons: Requires an MCP-compatible client to supply model context. Adoption needs a client configuration change per environment. Assistant’s internet requirement may still expose model calls externally.

Pros: Operates locally, keeping source code on the developer’s machine. MCP-native integration lets AI clients access project structure directly. Detects Go-specific concurrency risks and measures cyclomatic complexity. LRU cache reduces latency for repeated analyses in active sessions.

Cons: Static analysis outputs require developer validation before changes. Requires an MCP-compliant host and Node.js for npm installation option. Local processing depends on the machine’s resources for very large repositories.

Pros: Passive recording captures network, console, DOM, and screenshots for post-mortem analysis. DAP support enables breakpoint-level debugging across six programming languages. Framework-aware tracking offers component-level context for React and Vue. Acts as an MCP server and CLI for agent integration.

Cons: Diagnosis depends on completeness of recorded browser sessions. Privacy and retention model not specified for uploaded session data. Requires environments that support the Model Context Protocol.

Pros: Executes SOQL queries and anonymous Apex from MCP clients. Uses local Salesforce CLI authentication, does not store credentials. Open-source codebase allows auditing and custom extensions. Manages org connections and CRUD operations via natural language.

Cons: Requires Node.js v18 or higher and Salesforce CLI installed. Depends on an MCP-compliant client for AI integration. CLI-first approach requires developer familiarity and setup.

Pros: Allows AI to create, read, update, and delete WordPress content.. Media library management available to connected AI assistants.. Provides site diagnostics including active plugin lists and health status.. Supports safe SQL query execution for advanced data retrieval..

Cons: Requires an MCP-compatible client such as Claude Desktop.. Administrative deletions are possible if permissions are granted.. Initial setup needs MCP client configuration and token issuance..