Theo Browne presents three Markdown-based skillsAn AI agent skill is a reusable package of instructions, resources, and tool guidance for performing a bounded kind of work. built from recurring work in his development workflowWorkflow automation uses software or AI to complete a connected sequence of routine steps with less manual effort.. Theo Browne uses Sol Loop for bounded model reviewsAI-assisted code review uses AI to inspect software changes for defects, risks, standards, missing tests, and opportunities for improvement while human reviewers retain responsibility., limits follow-up rounds to changed code and unresolved findings, and cautions against expanding scope to satisfy every reviewer suggestion.
Theo Browne describes Full Send as explicit authorization to completeAgent autonomy is the degree to which an AI agent can choose and execute actions without immediate human direction., verify, review and merge a pull request when defined gates pass. Theo Browne describes Full Stop as the same preparation with merging reserved for a personHuman-AI collaboration combines human judgment and accountability with AI speed, generation, analysis, and tool use in a shared workflow., and explains why direct reproduction and observed runtime evidence matter alongside automated checks.
Theo Browne recommends encoding the limits of trust in clear triggers, verification steps, escalation rules and recovery practices. Theo Browne’s personal merge statistics and model comparisons are not independent performance evidence. Sponsor segments, course promotion and subscription invitations are omitted; the catalogue does not endorse bypassing another service’s access rules.
Watch on YouTube
