synathic: an MIT-licensed SDK that verifies agent writes in Postgres instead of trusting the 200 OK
Gallegos_Daniel · reddit · 2026-09-17
Developer GallegosDaniel shares a common production pain: agents report success with clean logs, but the database never actually changed. In interviews he found the pattern everywhere — one person had 40 bookings confirmed to users that never landed in the table; the real cost was the uncertainty window where nobody knew which records were reliable.
He built synathic, an MIT-licensed SDK that wraps agent functions with a decorator and checks Postgres directly after execution against declared postconditions (e.g. "this row should exist with this field matching") — deterministic verification, not an LLM judging its own output. The key design is causality checking: a plain rowexists check passes even if the write failed on a pre-existing row, so the default behavior anchors a timestamp before the function runs and requires the row's updatedat to be newer, proving this execution caused the current state. He admits concurrent writers touching the same row remain an open gap.
A sync mode covers user-facing actions like payments and bookings — confirming the write landed before responding — while everything else defaults to fire-and-forget with no added latency. Postgres-only for now, and he's looking for production users to stress-test it.
More from coding & agent
- Code-first, UI on top: building bespoke brand design tools with AI — floguo · 2026-09-17
- Study of 7 models across Claude Code, Codex, Pi: harness barely affects success but swings cost — DavideCrapis · 2026-09-17
- AI trading bot built with Jev is down 85%, owner shrugs it off — generativist · 2026-09-17
- Redditor's 3-Day SoL-Pi Test: Memory Objects Save ~12k Tokens Per Tool Run — Garblyx · 2026-09-17
- Reviewing AI code through Steve Jobs' lens: unseen internals deserve beauty too — sergeykarayev · 2026-09-17
- Telling Gemini's Astra to log build and test times made it cut its own bottlenecks — yacineMTB · 2026-09-17