Erebus

Penetration testing tools for the CodeAgent — network scanning, service enumeration, and vulnerability checks

Obsidian Local REST API tools for the CodeAgent — manage vault notes, search, and metadata

Frida dynamic instrumentation tools for the CodeAgent — mobile app analysis and SSL unpinning

DuckDB in-process analytics tools for the CodeAgent — SQL on CSV/Parquet/JSON without a server

Supabase client tools for the CodeAgent — database CRUD, auth, storage, and edge functions

OSINT reconnaissance tools for the CodeAgent — domain, IP, email, and social media intelligence

Use BEFORE any creative work — creating features, building components, adding functionality, or modifying behavior. Explores user intent, requirements, and design before implementation.

Use when completing tasks, implementing major features, or before merging — request a thorough review that checks spec compliance, code quality, security, and tests.

Use when facing 2+ independent tasks (test failures, bugs, subsystems) that can be worked on concurrently without shared state or sequential dependencies.

Use when you have a written implementation plan to execute, especially in a separate session or when subagents are unavailable — load plan, review critically, execute all tasks, verify at checkpoints.

Use when implementation is complete and all tests pass — guides completion of development work by verifying tests and presenting structured options for merge, PR, or cleanup.

Use when receiving code review feedback, before implementing suggestions — requires technical rigor and verification rather than performative agreement or blind implementation.

Use when executing implementation plans with independent tasks — dispatch fresh subagent per task with two-stage review (spec compliance, then code quality) for high quality, fast iteration.

Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes — four-phase root cause investigation: find cause, find pattern, form hypothesis, implement fix.

Use when implementing any feature or bugfix, before writing implementation code — enforces RED-GREEN-REFACTOR cycle: write failing test first, then minimal code to pass, then refactor.

Use when starting feature work that needs isolation from the current workspace, or before executing implementation plans — creates isolated git worktrees with safety verification.

Use before claiming work is complete, fixed, or passing — run verification commands and confirm output before making any success claims. Evidence before assertions, always.

Use when you have a spec or requirements for a multi-step task, before touching code — creates comprehensive implementation plans with bite-sized tasks, exact file paths, and complete code in every step.

Read, search, create, and manage notes in an Obsidian vault via the Local REST API.

Build and maintain a persistent interlinked markdown knowledge base, optionally stored in an Obsidian vault.

Create new custom skills and save them to the user-skills directory so the agent can learn and use them in future sessions.

Send notifications to users via configured apprise channels (Telegram, email, Slack, etc.)

Control Philips Hue lights, rooms, and scenes via the OpenHue CLI — brightness, color, temperature, and scheduling.

Delegate coding tasks to Claude Code agent for refactoring, implementation, PR reviews, and complex operations.

Delegate coding tasks to OpenAI Codex CLI agent for one-shot tasks, background operations, and batch processing.

Provider-agnostic AI coding agent with TUI — one-shot tasks, interactive sessions, and parallel work patterns.

Generate ASCII art using pyfiglet, cowsay, toilet, image-to-ASCII conversion, and LLM-powered freehand art.

Create hand-drawn style diagrams using Excalidraw JSON format with shapes, arrows, and text labels.

Create interactive generative art and visualizations using p5.js with 2D/3D rendering, particles, and shaders.

Create webhook subscriptions for event-driven agent activation from GitHub, GitLab, Stripe, and CI/CD systems.

Full IMAP/SMTP email client via Himalaya CLI — list, read, search, compose, reply, and forward emails.

Find nearby restaurants, cafes, bars, and services using OpenStreetMap — no API keys required.

CLI bridge for discovering, calling, and managing MCP servers directly from the terminal using mcporter.

Connect to MCP (Model Context Protocol) servers for external tool integration. Configure stdio, SSE, or HTTP transports.

Fetch YouTube video transcripts and transform them into chapters, summaries, blog posts, or threads.

Interact with Gmail, Calendar, Drive, Contacts, Sheets, and Docs via Python with OAuth2 authentication.

Search and retrieve academic papers from arXiv. Query by topic, author, or ID. Generate BibTeX citations.

Monitor blogs and RSS/Atom feeds for updates. Discover feeds, parse entries, and track changes.

Plan mode for writing structured markdown implementation plans without executing code.

Read, write, edit files and execute shell commands to help with coding tasks.