Project configuration
Rules, AGENTS.md, slash commands, skills, and subagents so agents follow the repo’s standards — whether that’s Cursor, Claude Code, Codex, or another runtime.
Workflows
I ship with traditional full-stack workflows that follow proven engineering best practices — clear architecture, testing, reviews, and durable delivery habits — and layer agentic workflows on top when they amplify that craft.
Event-driven & always-on
Automations and always-on agentic workflows triggered by configured events — useful when work should continue without a constant interactive session.
Agents in products
Embedding agents into web apps with SDKs and APIs — Cursor SDK and similar surfaces — when the product needs agent capabilities, not just the editor or CLI.
MCP tool access
Configuring MCP servers so agents can reach databases, browsers, internal APIs, and other tools beyond the local codebase.
Institutional knowledgebases
Curating durable project docs, decision records, and domain references so agents inherit shared context instead of rediscovering it each session.
Hooks & quality gates
Wiring hooks and gates so agents hit lint, tests, or policy checks before and after turns — enforcement that keeps institutional standards from being optional.
Nocciolo
Italian for kernel / core — the durable core of project knowledge that agents inherit.
AI agents start every session cold — re-learning architecture decisions, coding standards, domain invariants, and the why behind the work from scattered docs and tribal knowledge. That wastes tokens, produces inconsistent output, and breaks long-running agentic workflows.
I’m building Nocciolo as a company brain configuration layer: it turns durable project knowledge — ADRs, READMEs, decision records — into a properly configured memory bank agents can retain and recall. By seeding banks likeHindsight, agents inherit shared context instead of rediscovering it every session.
- Knowledge extraction — scans repos for durable knowledge while ignoring ephemeral noise.
- Contextual seeding — configures missions and directives for memory banks, tuned for software projects.
- MCP integration — emits Model Context Protocol snippets to wire memory into Cursor, Claude Code, and other agent harnesses.
- Local-first governance — keeps sensitive architecture decisions under team control, versioned alongside the code.