Designing, evaluating, and shipping autonomous AI agents.
9632 posts
A collection of skills to turn claude code/codex into a long-lived personal tutor.
Safe, read-only Android state snapshots for AI coding agents via ADB.
Practical, battle-tested AI agent skills for local/offline scenarios: local-vision (LM Studio offline image understanding/OCR) + veusz-plot...
Open Agent Skill that keeps repository context, feature specs, change handoffs, and README summaries synchronized across AI coding tools.
A filesystem contract (.workflow/meta.json) + 39 agent skills that take a product from idea to production: web (Next.js 16) & mobile (Expo/R...
Codex Skill for building ST-compatible independent RP applications.
Portable marketing second brain for Codex, Claude Code, and Agent Skills with 502 sources for strategy, SMM, content, growth, CRO, and buyer...
HMAP - Human Maintainability Assessment of Programming Projects（人类可维护性评估 skill）
An agent-native protocol for iterative adversarial research and the most defensible evidence-bound conclusion.
Three small, independent guardrails for AI coding agents: an anti-AI-slop writing checklist, a code-comment hygiene checklist, and a standin...
Continue work across Codex, Claude Code, and Cursor using local session transcripts.
Hypothesis-led consulting workflow from ambiguous brief to evidence, analysis, storyline, and finished deck
