Skills2026-06-28 · source-backed
Make your CI review gates structurally different, not redundant.
Story
One model reviewer, one deterministic linter or policy engine, one human on high-blast-radius paths. Seven correlated AI gates share blind spots and bill you for the privilege (Andrew Nesbitt). Cap agent-to-agent comment loops with a hard turn limit so two reviewers can't run up a five-figure inference bill arguing.
Related stories
Each link below shares sources, entities, or timing with this story.
Shared entity: Andrew Nesbitt / Same source / Shared topic
CVE-2026-LGTM: seven AI review gates, one malicious package, $41,255 in burned inference
Both cover Andrew Nesbitt; cite the same source (Andrew Nesbitt); overlapping topics (comment, correlated, different, inference, reviewer).
Shared entity: Andrew Nesbitt / Earlier coverage
Protestware Now Targets AI Coding Agents, and Your Build Output Is the Attack Surface
Both cover Andrew Nesbitt; earlier Andrew Nesbitt coverage from 2026-05-29.
Shared topic / Tension
GolemUI ships a declarative form engine
Overlapping topics (comment, engine); pushes against this story (vs).
Human-AI Code Review: Reviewers Need 11.8% More Rounds on AI-Generated Code
Overlapping topics (human, reviewer); pushes against this story (versus).
Andrew Ng: Agentic Reviewer Matches Humans.
Overlapping topics (human, reviewer); pushes against this story (vs).
Inside the M4 Apple Neural Engine (351 pts, 103 cmts)
Overlapping topics (engine, inference); pushes against this story (vs).
Same source domain
Willison Runs an Open Model That Writes Its Own Training
Reported by the same outlet (simonwillison.net).
shot-scraper 1.10 adds a command to record agent demos
Reported by the same outlet (simonwillison.net).
Source trail
Entities
Provenance
- Canonical issue
- Ramsay Research Agent — June 28, 2026
- AI generated
- no
- Story unit
- 2026-06-28-make-your-ci-review-gates-structurally-different-not-redundant
- Labels
- source-backed, canonical briefing excerpt