traits.sh turns agent code-review feedback into reusable traits, enforcing Redacted for secrets
samgoodwin89 · x · 2026-11-01
- samgoodwin89 added a trait on traits.sh requiring sensitive data like API tokens and potential PII to use Effect TS's Redacted type at application boundaries, so generated code automatically complies.
- traits.sh distills each correction into a reusable trait (name, subject, requirement, purpose, example diff); the agent asks which insights to keep, and a decision model checks future code against all approved traits, flagging violations.
- Quoted case: arjunkmrm found combining Cloudflare's extraction and decision models beats OpenAI's privacy filter at detecting private data in legal documents — "extraction + decision models together" looks like a generally good pattern.
More from coding & agent
- Dimillian's Codex goes off the rails, pulls up Backrooms images as game references — Dimillian · 2026-11-01
- Dev proposes cross-device agent session mesh over tailnet, tags OpenAI and Anthropic — MannyKayy · 2026-11-01
- Kanban board for your Codex/Claude Code agents: Multica tracks agent tasks as issues — jiayuan_jy · 2026-11-01
- If your AI scores 100%, your eval is wrong — an AI safety researcher's eval playbook — AI Engineer · 2026-11-01
- Control Vectors experiment: letting an agent manage its own activations — sloppenheimer · 2026-11-01
- FlashRSI open-sources RSI research: 12 algorithms, 30+ benchmarks, OOD gains still elusive — HuaxiuYaoML · 2026-11-01