Technical

A common point of confusion in Terraphim is understanding the difference between FFF (fzf-like fuzzy file finder) and Ripgrep haystack. This post clarifies what each does and how they complement each other.

This guide documents the step-by-step process for setting up a Frontend Developer role in Terraphim, including knowledge graph validation, MCP server configuration, and integration with OpenCode.

As AI agents write more and more production code, the question shifts from "can the agent write code?" to "can we verify the code is good?" Two tools tackle this from opposite ends: Sentrux measures structural health, Terraphim measures semantic completeness. Neither is complete without the other.

Vector embedding calls are the hidden tax on every RAG pipeline. You pay latency, you pay API cost, you get probabilistic results that vary run to run. There is a class of problem where none of that is acceptable. This post shows how Terraphim replaces embedding calls with Aho-Corasick finite-state automata -- deterministic, auditable, and under one millisecond for 1.4 million patterns.

We have published a comprehensive walkthrough showing how to build a specialised front-end developer agent using Terraphim's knowledge graph system, dual haystacks, and deterministic search. This post walks through the key concepts and links to the full guide.

Vector databases are probabilistic and slow. Graph embeddings are deterministic and sub-millisecond. If you are building context for an AI coding agent — or any system where you need to know why a result came back — the difference is not academic. It changes what your application is allowed to promise.

Your AI coding agent already has a knowledge graph. It is just not yours yet. The model knows GitHub, Stack Overflow, and the public training corpus -- it has no idea that in your project npm should be bun, that RFP is shorthand for acquisition need, or that the email about the Stripe receipt for the Obsidian licence lives in your Fastmail mailbox. This post shows the smallest path to fixing that for both Claude Code and opencode, using Terraphim and the three roles we have published over the last week (Terraphim Engineer, Personal Assistant, System Operator).

Two paths. CLI first.

AI coding agents are making us worse engineers, unless we add discipline back. Here is what we do instead of vibe coding, and how you can do it too in 30 seconds.

AI coding assistants are fast, productive, and occasionally catastrophic. One misplaced rm -rf, one accidental git reset --hard, and hours of uncommitted work vanish.

Jeffrey Emanuel (@Dicklesworthstone) built Destructive Command Guard (dcg): a Rust binary with SIMD-accelerated pattern matching, 49+ security packs, and a fail-open design. It is one of the best tools to come out of the AI agent safety space, and it solved a problem we had been fighting with regex hacks.

This post shows how we integrated dcg with OpenCode using its plugin hook system, so destructive commands are intercepted before they run.