hermes-agent-rtk-caveman

Integrates Caveman templating engine with RTK (Rust Token Killer) for token-optimized output formatting. Provides a wrapper script and patterns for using Caveman templates to compress CLI output.

Optimize Hermes context usage — expand RTK coverage, prune sessions, manage memory, set up auto-prune cron. Use when token usage is high or RTK coverage is low.

Set up SSH authentication for custom Git hosts (non-standard ports, custom domains, VPN-dependent hosts) with verification workflow and troubleshooting for common connectivity issues.

Manages MySQL services via Homebrew with RTK integration for token optimization. Provides easy start/stop/restart/status commands and monitoring capabilities.

Install and configure Syncthing for folder sync between macOS and Android devices. Covers brew install, service setup, web UI automation for folder creation, device ID extraction, and cross-device pairing instructions. Use when user needs to sync folders between Mac and phone/tablet.

Synchronize Zenith Finance ledger and transactions to Google Sheets with automatic updates. Creates/maintains a spreadsheet with Dashboard (summary, charts) and History (full transaction log) sheets.

Maintain Zenith Finance local files (Ledger.md and Transactions.md) in an easy-to-maintain format. Provides simple key-value format for ledger, standard transaction format, helper scripts, and procedures for fixing common formatting issues while maintaining compatibility with Google Sheets sync.

Advanced Finance & Asset Manager. Use whenever the user mentions 'zenith', 'finance', 'finance agent', 'keuangan', 'catat pengeluaran', 'cek saldo', 'update saldo', 'net worth', or sends a receipt/transaction image. Handles multi-account ledger, transaction logging, OCR receipt processing, and budget alerts.

Auto-sync Zenith Finance data (transactions, ledger) to Google Sheets. Creates/updates a spreadsheet with Dashboard and History sheets.

Set up GitHub authentication for the agent using git (universally available) or the gh CLI. Covers HTTPS tokens, SSH keys, credential helpers, and gh auth — with a detection flow to pick the right method automatically.

GitHub setup with SSH keys and understanding when tokens vs SSH are needed. Covers creating dedicated SSH keys, configuring SSH config, and the limitations of SSH vs gh CLI token requirements.

Read, search, create, and organize notes across multiple Obsidian vaults. Supports smart vault/folder resolution from natural commands. Shorthand trigger: "obi" — treating any mention of "obi" the same as "obsidian". Use when user wants to save/create notes, organize by category/folder, manage vaults, or references any Obsidian-related action. Triggers on: "obsidian", "obi", "catatan", "notes", "save ke vault", "masukin ke obsidian", "buat notes", vault paths, or any note creation request.

Maintain and fix inconsistencies in documentation indexes, tables of contents, and structured markdown files

Properly set up Google Workspace OAuth2 authentication for Hermes agent, handling Python environment and module dependencies.

Gmail, Calendar, Drive, Contacts, Sheets, and Docs integration via Python. Uses OAuth2 with automatic token refresh. No external binaries needed — runs entirely with Google's Python client libraries in the Hermes venv.

Use this skill any time Adit wants to create, update, search, or manage notes in Obsidian vault. Also triggers when the user says "obi", "add notes", "catat di obsidian", "simpan notes", "buat note", or mentions saving any text/ideas/snippets to Obsidian. This is a general-purpose note management skill — not limited to any specific topic. Can create notes in any folder/subfolder structure within the vault. Supports creating notes from clipboard, typed text, code snippets, meeting notes, ideas, TODOs, and more. If the user says "obi" followed by any text, use this skill immediately.

Create automated financial analysis skills with scheduled execution and result storage. Use when you need to build custom stock/scanning analysis that runs automatically and saves results to Obsidian for daily review.

Karpathy's LLM Wiki — build and maintain a persistent, interlinked markdown knowledge base. Ingest sources, query compiled knowledge, and lint for consistency.

End-to-end pipeline for writing ML/AI research papers — from experiment design through analysis, drafting, revision, and submission. Covers NeurIPS, ICML, ICLR, ACL, AAAI, COLM. Integrates automated experiment monitoring, statistical analysis, iterative writing, and citation verification.