One malformed item killed my agent pipeline — now loading repairs it instead
petrusenko_max · x · 2026-10-31
Two runs of the author's WhatsApp sender crashed on the same day — not on bad messages, but on the shape of the plan file (a KeyError on missing checkedafter, a ValueError on an event id with no date).
His fix: both the sender and builder now repair the plan at load time —
- fill missing checkedafter and keep running
- fix bad notbefore values instead of raising
- fill missing jid, phone, kind, key fields
- add an eventdate to undated event ids
- reject only unrepairable items, logging every rejection so nothing vanishes silently
- run the same repair in both load paths so neither trusts the other
Principle: a malformed item used to take down the whole run; now it costs one item.
More from coding & agent
- Letting Claude edit an entire YouTube video, results shared — prasenx · 2026-10-31
- PotionUI 0.0.18: open-source self-hosted studio for AI image, video and music on your own GPU — 0roborus_ · 2026-10-31
- Vals Code Migration Leaderboard: Claude Sonnet 5.5 Tops at 86.88%, DeepSeek Costs $1.76 — davidbau · 2026-10-31
- NetHack port challenge: AI transpilers hit 99.5% correctness but tank readability — davidbau · 2026-10-31
- Rex's floating panes demo sparks take that Mitchell Hashimoto is building an agentic meta-OS — dankaplan · 2026-10-31
- LLM judges are strongly biased toward position, researcher warns to randomize everything — antoine_chaffin · 2026-10-31