https://egoistai.com/articles/context-window-truncation-safeguardsEgoistAIen2026-09-07T00:00:00.000ZContext Window Truncation Safeguards: Stop AI Agents From Forgetting Critical Instructionshttps://egoistai.com/articles/mcp-oauth-security-ai-agentsEgoistAIen2026-09-07T00:00:00.000ZMCP OAuth Security for AI Agents: Scope Tokens, Bind Resources, and Block Confused Deputieshttps://egoistai.com/articles/structured-output-repair-loopsEgoistAIen2026-09-07T00:00:00.000ZStructured Output Repair Loops: Recover Invalid LLM JSON Without Hiding Failureshttps://egoistai.com/articles/ai-agent-state-checkpointingEgoistAIen2026-09-06T00:00:00.000ZAI Agent State Checkpointing: Resume Long Tasks Without Repeating Side Effectshttps://egoistai.com/articles/llm-timeout-budget-propagationEgoistAIen2026-09-06T00:00:00.000ZLLM Timeout Budget Propagation: Stop Nested AI Calls From Outliving the User Requesthttps://egoistai.com/articles/policy-as-code-ai-agent-guardrailsEgoistAIen2026-09-06T00:00:00.000ZPolicy as Code for AI Agents: Turn Safety Rules Into Testable Runtime Controls