Posts

A mini agent gateway is the code layer you place between your LLM and its tools that constrains what the model can say, call, and spend. Instead of trusting a raw model endpoint, you add schema-based argument validation, a tool allow-list, pre-execution guardrails, and model routing so every model call is checked before it touches a real system. The result is an LLM you can actually run in production: cheaper, safer, and deterministic. ...

Frugal Tokens is a free, local-first, read-only tool that lets you see exactly how many tokens each of your AI coding agents consumed and what that cost you, across OpenCode, Claude Code, PI, Codex, and Cursor. It runs locally in Deno, never touches your agent databases, and surfaces per-session token usage plus the hidden cost of broken prompt-cache reuse. It is the sharpest per-session cost diagnostic we tested for teams that want to measure ROI on AI coding spend. ...

A universal memory protocol is an open, standardized format for storing AI agent memory so that any agent or tool can read and write it, regardless of which vendor built it. Today most agent memory is trapped in proprietary silos: each tool stores context in its own schema, so switching agents means losing your history. A universal memory protocol is the missing “Rosetta Stone” that makes context portable, interoperable, and portable between tools. ...

PMB is a local-first, MCP-native memory layer for AI coding agents that stores everything in one SQLite file on your disk and — unusually — measures whether that memory actually changes outcomes. Instead of claiming “+X% faster,” it scores each surfaced lesson against the turn’s real result, so you can see when memory is genuinely helping. Why Coding Agents Forget — and Why “Memory” Is the Hard Part Every coding agent you’ve used — Claude Code, Cursor, Codex, Windsurf, Zed, VS Code, gemini, opencode, continue — starts each session with a clean slate. The model has no persistent recollection of the architecture decisions you made last week, the test suite you just fixed, or the deployment gotcha that cost you an afternoon. This is the “forgetting problem,” and it is the single biggest reason long-running projects feel like they restart from zero every time you open a new session. ...

ContextNest vs Mem0 vs Zep is the wrong framing: these three are complementary layers of the same memory stack, not competing products. Zep provides session log memory, Mem0 handles personalization memory, and ContextNest governs corporate knowledge — and production agents typically need all three. A single memory database is the most common architectural pitfall, because it cannot serve session continuity, user personalization, and compliance-ready knowledge at the same time. This guide dissects each layer, compares them head-to-head, and shows how to stack them. ...

Open-source memory for coding agents synced over SSH lets you carry your AI assistant’s context, preferences, and project knowledge across every machine you work on — without a cloud service. The most portable approach is a git-backed markdown store: symlink your agent’s memory files, push and pull over SSH, and let auto-sync hooks handle the rest. This guide walks through the three proven methods and how to pick the right one. ...

MCP observability turns the Model Context Protocol into a two-way interface: AI agents don’t just call tools, they receive ground-truth telemetry from kernel tracepoints, eBPF programs, and kprobes. By exposing low-level system instrumentation through MCP servers, agents get a real-world model of the live system instead of hallucinated state — closing the observability gap that traditional APM leaves wide open. What Is MCP and Why It Needs Observability The Model Context Protocol (MCP) is an open standard that standardizes how AI agents discover and invoke tools, resources, and prompts. Instead of every agent building bespoke integrations with every service, MCP defines a common protocol: a host (the agent runtime) connects to MCP servers, which expose tools the model can call and resources it can read. ...

Heimdall is an open-source, CPU-only knowledge layer that gives AI coding agents a verified, self-healing memory across every repository you work in. Instead of returning plausible-but-unverified matches, every search result carries a trust verdict — STRONG, WEAK, REBUILT, or STALE — re-checked against your live filesystem at query time. It indexes with tree-sitter and local embeddings, spends zero tokens on memory maintenance, and re-anchors moved files automatically. What is Heimdall and why does it exist? Heimdall is a trust-verified knowledge layer built specifically for AI coding agents. It was released on Hacker News on August 22, 2026, and is published on npm as @ariantdeva/heimdall at version 0.2.1. The project’s core claim is simple: most agent memory tools return results that look right but have not been checked against the actual state of your codebase. Heimdall exists to close that gap. ...

A project wiki is a durable, versioned set of instructions and context that AI coding agents load into every session, so they never start from zero. The open AGENTS.md format — used by over 60,000 open-source projects — and Claude Code’s CLAUDE.md files give agents a predictable place to find build steps, conventions, and architecture. Because these files live in git, they are traceable: every change is recorded, reviewable, and honest. This guide shows you how to build one. ...

Simmis is a self-hosted workspace where people and AI agents share a single, versioned, queryable memory — chat rooms, wiki pages, knowledge bases, a double-entry ledger, and a code repository all live on one substrate instead of being scattered across siloed tools. Built in Clojure on the replikativ stack, it treats humans and agents as the same kind of participant, so @mentions, governance, and history work identically across both. It is an early, MIT-licensed “release early” project that is used daily but still has rough edges. ...