Discover +384 AI Coding apps & tools

Pros: Provides live crates.io lookups for assistants. Reads local project structure for context-aware suggestions. Integrates with Cargo for dependency-aware responses.

Cons: Requires an MCP-compliant client to operate. Internet required for external crate searches. Functionality is limited to the Rust ecosystem.

Pros: Local-first processing, no cloud uploads or API keys required. Hybrid search combining BM25, vector embeddings, and PageRank. Blast-radius refactoring analysis (sverklo_impact) for PR risk assessment. Supports 12+ languages via tree-sitter incremental parsing.

Cons: Requires Node.js 24 or higher on the host machine. Initial repository indexing consumes local CPU and storage. Agent outputs still require human verification for high-risk decisions.

Pros: Automates Master Code detection, removing manual hexadecimal searches. Parses Action Replay, GameShark, and CodeBreaker into PNACH. Batch processing for handling multiple code strings at once. Portable Windows tool with no complex installation required.

Cons: Windows-only, requires a .NET-compatible runtime. Simple GUI may lack advanced code-editing controls. Users must understand PNACH usage to apply patches correctly.

Pros: Enables AI to perform local file edits within defined project scope. macOS-optimized runtime for developers working on local projects. MCP compatibility integrates with clients like Claude Desktop and Cursor. TypeScript codebase simplifies extension for JavaScript teams.

Cons: Requires Node.js and an MCP-compatible client for operation. macOS-first optimization limits out-of-the-box parity on other systems. Generated code edits still need human review before merging.

Pros: Exposes file keys and node IDs for element-level model access. Supports multi-file workflows for cross-document design references. Proxies requests through the plugin to avoid REST API rate limits. Open-source project from Gethopp, with developer tooling experience.

Cons: Requires an MCP-compatible client and local server setup. Installer and runtime require Node.js and local network access. Generated outputs need human review before production use.

Pros: Exposes Xcode project structure to MCP-compatible AI models. Runs builds and returns diagnostic errors and warnings to clients. Executes unit and UI tests and reports outcomes to the assistant. Open-source codebase enables public review and community contributions.

Cons: Requires macOS with Xcode and command line tools installed. Command-line server needs manual configuration with an MCP client. Primary focus on .xcodeproj/.xcworkspace, limited package-only focus. Automated file modifications require human verification before merging.

Pros: MCP compliance enables straightforward integration with agent clients. Supports file read/write, directory navigation, and workspace search. Allows executing shell commands for end-to-end edits and tests. Open-source repository available for inspection and contribution.

Cons: Local command execution requires strict user supervision. Depends on a Node.js environment and an MCP client. Targeted at early adopters familiar with agent workflows.

Pros: Open-source codebase allows community review and contributions. Supports Sublime Text 3 and 4 on Windows, macOS, and Linux. Exposes editor content and project metadata to MCP workflows.

Cons: Requires an external MCP-compliant server to function. Server connections configured via JSON, needing manual edits. No bundled AI model; models run on external servers.

Pros: Injects idiomatic guidance into the model context through MCP. Queryable tenets let agents request specific, language-tailored style guidance. Installs and runs with common Python tooling such as uv or pip.

Cons: Improves style but does not ensure semantic correctness. Currently limited to included philosophies, e.g., Python and Go. Requires an MCP-compatible client and Python runtime.

Pros: Protocol-native evaluation interface compatible with MCP hosts. Produces numeric scores with explanatory qualitative reasoning. Provider-agnostic design supports multiple backend LLMs. Exposes judging as callable tools for autonomous agents.

Cons: Evaluation quality depends on the chosen backend LLM. Requires Node.js environment and MCP host configuration. Targeted at developers, not non-technical users.

Pros: MCP-native interface for agent-driven web actions. Uses Chromium rendering for reliable JavaScript-heavy page handling. Produces HTML, DOM extracts, and high-resolution screenshots. Quick run via npx for fast experimentation.

Cons: Requires an MCP host and a Node.js environment to operate. Search provider integrations may need environment variables. Targeted at developers rather than nontechnical end users.

Pros: Enables byte-range reads so models access specific segments of large files. Written in Go, offering low resource overhead when streaming files. Runs locally as an MCP server, keeping files off third-party cloud storage. Compatible with any MCP host, including Claude Desktop.

Cons: Requires MCP host and manual configuration, challenging for non-technical users. Search results are most effective on UTF-8 text, limited on binary files. Model interpretations of returned bytes require human verification.

Pros: Outputs Go Abstract Syntax Trees for machine-readable code representation. MCP-native, enabling direct connection with MCP-capable assistants. Read-only local file access preserves source integrity during parsing. Go-based server implementation reduces query latency for interactive sessions.

Cons: Requires the Go toolchain to resolve dependencies. Needs an MCP-capable client to be useful in workflows. Specialized to Go, not suited for multi-language repositories. Parsing completeness depends on local module resolution.

Pros: Mode management centralizes instruction state for repeatable assistant behaviors. Instruction library enables persistent, reusable prompts across sessions. Local stdio server deployment supports host-side data control. Programmatic APIs allow scripted mode changes and integration.

Cons: Requires an MCP-compatible host such as Claude Desktop or VS Code. Setup requires Python and MCP extension familiarity. Persistence depends on host implementation and configured storage.