Chapters
DOC-05 / Technical reference
Technical Documentation
The harness of Synedre, in depth. A public, sanitised reference: how the data layer, the agentic core, chantiers, automates, the hub, email, memory, deployment and façades actually fit together.
11 chapters
Anatomy of Shyrka — see the body firstNightly loop
The system maintains its own documentation
Every night, the harness inspects its own state — doc↔code drift, health check, gated regeneration. This terminal shows the actual log of the last session.
Chapters
Agentic harness overview
The harness transforms a request (email, console, cron) into a deployed action by chaining intent classification, pseudo-TTY LLM spawn, two-level quality validation, and a learning loop — with no business state stored outside the database.
The Data Layer
Describes the Synedre OS data architecture: a single PostgreSQL database (un composant interne/un composant interne), the three access paths (Nuxt/Nitro, agentic Python, Drizzle ORM DDL), the table families by prefix, and the associated naming conventions.
The Agentic Core: Atlas and the Agents
Describes the architecture of the agentic engine: LLM classification of incoming emails, headless spawning of Claude Code via pseudo-TTY, injection of agent personas, and post-spawn orchestration (deploy, QA, recap).
Worksites, Jobs & Tasks — Data Model and API
Describes the worksite/job/task hierarchy in Synedre OS, the DB data model, canonical statuses/scopes, and the associated Python API.
Automations, crons & runs
Describes the non-conversational execution layer of Synedre OS: the 209 Python facades, their cron wrapper, the un composant interne registry, and the boundary between run doctrine and agent unit.
The Hub (/hub/*)
Presentation of the Synedre OS private administration interface (mothership-app), its Nuxt layered architecture, the map of its modules and pages, as well as the authentication and session management system.
Inbox hub & Atlas Inbox — two email pipelines
Describes the two IMAP→DB pipelines of the harness (contact@ sorted for the hub, atlas@ triggering agentic actions), the outbound sending facade, and the associated blocking doctrines (AV scan, zero direct client contact).
Memory & Learning — Three-Level Architecture
This chapter describes the three memory levels of the Synedre OS harness (reflex, Zettelkasten, vector pgvector), the associated indexing automata, and the loop for capitalizing errors into reusable rules.
Deployment & infrastructure
Describes the Synedre OS release pipeline: asymmetry between ./deploy (AI, preprod) and ./ship (Alex, production), the YAML dispatcher, the build-host pattern without build VPS, and the associated rules (secrets, commit-before-deploy, drift, smoke).
Catalogue of Synedre OS Facades & Entrypoints
Inventories and describes all executable entry points of the agentic harness (synedre/*.py and bin/*) with their family, role, and invocation mode, enabling an engineer to quickly navigate the repository.
Skills, agents & hooks — Synedre OS agentic harness
Describes the complete wiring of the synedre-os Claude Code harness: available skills, delegatable sub-agents, and event hooks that enforce the commit-in-flow doctrine, prod/email guardrails, and memory injection.