Lab: AI Security and Tool Analysis

2026-08-11 | Agents Running Agents: Codex Meets OpenCode and DeepSeek A hello-world DeepSeek request became a safer agent-to-agent runner, a dated route preflight, and a Cost per Accepted Result protocol.
2026-06-05 | LLM App-Hacking Field Test: $1,500 Takeaways A cautious read of Kasra Rahjerdi's informal $1,500 field test on whether LLM agents could exploit a deliberately vulnerable app.
2026-02-11 | Stack Snapshot — 11 Feb 2026 Point-in-time record of tools, models, and infrastructure as of 2026-02-11. Kimi quota resolved, Codex 5.3 solid, Opus 4.6 deadline mode analysis.
2026-02-09 | Stack Snapshot — 9 Feb 2026 Point-in-time record of tools, models, and infrastructure as of 2026-02-09. Kimi K2.5 dominant, Flow/Veo3 testing, rate limit reality documented.
2026-02-04 | Stack Snapshot — 4 Feb 2026 Point-in-time record of tools, models, and infrastructure as of 2026-02-04. OpenCode limits reduced, Kiro CLI changes, early MCP experiments.