ECOSYSTEM
Six tools. Each solves a distinct problem. All six converge on the same AI agent — giving it memory, structure, efficiency, skill, knowledge, and creative power at once.
INTEGRATION
The Knowledge Pipeline
obsidian-forge runs as a daemon, continuously organizing your Obsidian vault — architecture decisions, research, conventions — and strengthening its knowledge graph. When a note is ready, `alcove promote` imports it into alcove's indexed project docs. From that point, any MCP-compatible agent can query it on demand: BM25 full-text search, vector embeddings, hybrid ranking. Personal notes become authoritative, searchable project context.
The Invisible Optimizer
llm-transpile integrates directly with the agent's file-reading pipeline. Install it once and it hooks into Claude Code's PostToolUse event — every `.md`, `.html`, or `.txt` file an agent reads is automatically compressed into a structured bridge format before it lands in the context window. No prompting required. The agent reads the same information, uses fewer tokens, and reaches further into complex tasks without hitting limits.
The Agent That Learns
epic-harness replaces 30+ slash commands with 6 and adds a Ring 0 hook system that observes every tool use, detects failure patterns, and reflects them into a cross-project SQLite knowledge graph. This memory is independent of alcove — it tracks decisions, resolutions, and patterns across sessions, not documents. The result: an agent that improves over time, auto-triggers the right skill for the current context, and can evolve new skills from its own failure history.
Engineering Wisdom On Demand
When the agent faces a design decision or code smell, Episteme's MCP server surfaces the relevant design pattern, law, or refactoring technique from its knowledge graph — 100+ canonical entities linked by semantic relationships. The agent doesn't just guess; it grounds recommendations in proven software engineering principles. Works offline, zero config.
Long-Form Creation Pipeline
Velith turns the agent into a publishing engine. Six phases — onboarding, ideation, outlining, drafting, editing, publishing — each with quality gates. Seven genre templates and seven specialized agents handle different aspects. The result: structured, coherent long-form content from blank page to EPUB/PDF.
When All Six Converge
In a single agent session all six are active simultaneously. The agent reads a file — llm-transpile compresses it. It needs project background — alcove surfaces the right docs. It faces a design decision — Episteme grounds it in proven patterns. It chooses a strategy — epic-harness dispatches the proven skill. It creates long-form content — Velith orchestrates the full publishing pipeline. The knowledge behind those docs was shaped by obsidian-forge weeks earlier. Six tools, one agent, compound intelligence.
THE SIX TOOLS
Obsidian vault automation daemon. Generates notes, strengthens the knowledge graph, auto-commits, and runs headlessly as a single binary. Feeds alcove via `alcove promote`.
MCP server for on-demand project docs. Stores architecture decisions and conventions; serves them via BM25 + vector hybrid search to any MCP-compatible agent.
Hooks into the agent's PostToolUse event. Every file read is auto-transpiled into a token-efficient bridge format — same fidelity, fewer tokens, no agent configuration needed.
6 commands replace 30+. Ring 0 hooks observe, reflect, and build a SQLite pattern memory across sessions. Detects failure patterns and evolves new skills automatically.
Offline-first knowledge graph connecting design patterns, refactoring techniques, and software laws through semantic relationships. Built for AI agents — surfaces engineering wisdom on demand via MCP.
AI-native publishing system with 6-phase pipeline, 7 genre templates, and 7 specialized agents. From blank page to publishable EPUB/PDF with quality gates at every stage.