Personal blog about ML infrastructure, self-hosted AI, distributed systems, and technical debugging. https://anyech.github.io/jingxiao-cai-blog/ en-us Tue, 12 May 2026 02:57:36 +0000 A technically true row-count alert became an alert-tuning lesson: record weak proxy crossings, but page only when they combine with byte pressure, latency pressure, cleanup pressure, or user-visible impact. https://anyech.github.io/jingxiao-cai-blog/true-alert-wrong-page-agent-ops.html https://anyech.github.io/jingxiao-cai-blog/true-alert-wrong-page-agent-ops.html Mon, 11 May 2026 00:00:00 +0000 A degraded coding-agent lane is not automatically a local repair task. Classify it as passive watch, upstream wait, or a narrow adapter fix before changing fallback policy. https://anyech.github.io/jingxiao-cai-blog/coding-agent-route-drift-without-premature-fixes.html https://anyech.github.io/jingxiao-cai-blog/coding-agent-route-drift-without-premature-fixes.html Sun, 10 May 2026 00:00:00 +0000 A daily scan job generated its report, but the final delivery side effect failed. The fix was to replay the saved artifact, not rerun the whole workflow. https://anyech.github.io/jingxiao-cai-blog/when-report-exists-but-delivery-failed-agent-ops.html https://anyech.github.io/jingxiao-cai-blog/when-report-exists-but-delivery-failed-agent-ops.html Sat, 09 May 2026 00:00:00 +0000 A generated Markdown report failed over one heading-level jump; the durable fix was testing each rendered output surface as its own artifact contract. https://anyech.github.io/jingxiao-cai-blog/one-heading-level-broke-nightly-build-generated-rem-reports.html https://anyech.github.io/jingxiao-cai-blog/one-heading-level-broke-nightly-build-generated-rem-reports.html Fri, 08 May 2026 00:00:00 +0000 Detaching long-running agent work is not enough. The workflow also needs an admission gate, a validated return path, and audit feedback when coverage misses. https://anyech.github.io/jingxiao-cai-blog/long-running-agent-work-needs-bridge-back.html https://anyech.github.io/jingxiao-cai-blog/long-running-agent-work-needs-bridge-back.html Thu, 07 May 2026 00:00:00 +0000 A self-hosted agent-ops debugging story: raw SQLite can still see rows while the runtime registry restore fails, so reproduce on copies before touching production. https://anyech.github.io/jingxiao-cai-blog/sqlite-empty-corrupt-task-registries-without-touching-prod.html https://anyech.github.io/jingxiao-cai-blog/sqlite-empty-corrupt-task-registries-without-touching-prod.html Wed, 06 May 2026 00:00:00 +0000 A self-hosted agent-gateway performance lesson: if a tiny session-list request builds hundreds of rich rows before filtering, limit is too late to save you. https://anyech.github.io/jingxiao-cai-blog/10-second-session-list-prefilter-row-build-agent-gateway.html https://anyech.github.io/jingxiao-cai-blog/10-second-session-list-prefilter-row-build-agent-gateway.html Sat, 02 May 2026 00:00:00 +0000 A lightweight agent-operations pattern for closing external threads cleanly: make constraints explicit, record the decision, finish the action, and define reopen criteria. https://anyech.github.io/jingxiao-cai-blog/closing-external-threads-cleanly-agent-ops.html https://anyech.github.io/jingxiao-cai-blog/closing-external-threads-cleanly-agent-ops.html Fri, 01 May 2026 00:00:00 +0000 Why self-hosted AI-agent updates need production-deployment discipline: preflight, backup, staged rollout, human activation, adoption scans, and verification. https://anyech.github.io/jingxiao-cai-blog/ai-agent-updates-production-deployments-runbook.html https://anyech.github.io/jingxiao-cai-blog/ai-agent-updates-production-deployments-runbook.html Thu, 30 Apr 2026 00:00:00 +0000 Why design-tool integrations need capability gates before LLM generation: validate inputs, route readiness, model config, and artifact proof early. https://anyech.github.io/jingxiao-cai-blog/design-tool-integrations-capability-gates-llm-config.html https://anyech.github.io/jingxiao-cai-blog/design-tool-integrations-capability-gates-llm-config.html Thu, 30 Apr 2026 00:00:00 +0000 How I modernized agent skills with problem-first discovery, intake gates, thin wrappers, package hygiene, capability gates, and consolidation instead of skill sprawl. https://anyech.github.io/jingxiao-cai-blog/modernizing-agent-skills-without-growing-skill-jungle.html https://anyech.github.io/jingxiao-cai-blog/modernizing-agent-skills-without-growing-skill-jungle.html Wed, 29 Apr 2026 00:00:00 +0000 Why AI-agent tool launches should prove auth intent, isolate ambient credentials, check route readiness, and block before side effects when the launch contract is unhealthy. https://anyech.github.io/jingxiao-cai-blog/fail-closing-agent-launches-auth-readiness-gates.html https://anyech.github.io/jingxiao-cai-blog/fail-closing-agent-launches-auth-readiness-gates.html Wed, 29 Apr 2026 00:00:00 +0000 An OpenClaw stage-environment pattern for a small VPS: fail-closed testing, zero-production-secret bootstrap, detect-only catalog refresh, and a mock-to-real-to-higher-risk ladder. https://anyech.github.io/jingxiao-cai-blog/fail-closed-stage-environments-ai-agents-vps.html https://anyech.github.io/jingxiao-cai-blog/fail-closed-stage-environments-ai-agents-vps.html Wed, 08 Apr 2026 00:00:00 +0000 How I run a config-backed multi-model panel in OpenClaw: launch guards, watchdog-backed waits, bridge-back delivery contracts, and honest dissent safely. https://anyech.github.io/jingxiao-cai-blog/consult-panel-orchestration-openclaw.html https://anyech.github.io/jingxiao-cai-blog/consult-panel-orchestration-openclaw.html Fri, 03 Apr 2026 00:00:00 +0000 How I got OpenClaw local memory search working on a small ARM VPS, now with safer rollout, stricter active-memory canary gates, session-list fast paths, and source hygiene. https://anyech.github.io/jingxiao-cai-blog/local-semantic-memory-openclaw-arm-vps.html https://anyech.github.io/jingxiao-cai-blog/local-semantic-memory-openclaw-arm-vps.html Thu, 19 Mar 2026 00:00:00 +0000 Complete tutorial on OAuth 2.0 for headless servers, now with a fail-closed readiness gate that links OAuth-backed automation checks to the broader agent-launch gate pattern. https://anyech.github.io/jingxiao-cai-blog/vps-oauth-survival-guide.html https://anyech.github.io/jingxiao-cai-blog/vps-oauth-survival-guide.html Wed, 25 Feb 2026 00:00:00 +0000 Why AI cron reliability needs exact-exec drivers: artifact-first execution, explicit timeout budgets, and deterministic delivery instead of freeform agent prompts. https://anyech.github.io/jingxiao-cai-blog/ai-cron-jobs-exact-exec-drivers.html https://anyech.github.io/jingxiao-cai-blog/ai-cron-jobs-exact-exec-drivers.html Mon, 27 Apr 2026 00:00:00 +0000 My AI agent runs autonomous cron jobs every night—security audits, health checks, and documentation—now updated with the exact-exec driver lesson that prevents false-negative wrapper alerts from hiding real command success. https://anyech.github.io/jingxiao-cai-blog/nightly-build-security-audits.html https://anyech.github.io/jingxiao-cai-blog/nightly-build-security-audits.html Mon, 02 Mar 2026 00:00:00 +0000 How I built a declarative change propagation system for cron automation: manifest-driven updates, contract-derived documentation blocks, and validation that keeps desired state from quietly drifting — now with a stage-validation ladder from mock to real to a narrow higher-risk lane. https://anyech.github.io/jingxiao-cai-blog/declarative-change-propagation-cron-system.html https://anyech.github.io/jingxiao-cai-blog/declarative-change-propagation-cron-system.html Fri, 27 Mar 2026 00:00:00 +0000 A deep dive into malicious skills in AI agent platforms, now updated with the LiteLLM incident, the first named downstream victim report, LangChain/LangGraph vulnerabilities, exposed Ollama servers, and an approved-but-blocked lesson from Moltbook's unstable post-incident API surface. https://anyech.github.io/jingxiao-cai-blog/supply-chain-attack-ai-agents.html https://anyech.github.io/jingxiao-cai-blog/supply-chain-attack-ai-agents.html Mon, 23 Feb 2026 00:00:00 +0000 A clean OpenClaw upgrade passed startup checks but regressed under real use. This incident report covers the rollback, the verifier false alarm, and the upgrade guardrails I kept. https://anyech.github.io/jingxiao-cai-blog/openclaw-upgrade-rollback-runtime-regression.html https://anyech.github.io/jingxiao-cai-blog/openclaw-upgrade-rollback-runtime-regression.html Thu, 02 Apr 2026 00:00:00 +0000 Some OpenClaw config changes apply live. Others trigger gateway restarts. Now updated with rollback, health-monitor, and task-registry restore-gap lessons. https://anyech.github.io/jingxiao-cai-blog/gateway-restart-behavior-openclaw.html https://anyech.github.io/jingxiao-cai-blog/gateway-restart-behavior-openclaw.html Wed, 11 Mar 2026 00:00:00 +0000 When Gemini route health drifts, the hard part is not picking a random next model. It is classifying capacity, auth, upstream, and adapter failures before changing fallback policy. https://anyech.github.io/jingxiao-cai-blog/gemini-capacity-exhaustion-fallback-lanes.html https://anyech.github.io/jingxiao-cai-blog/gemini-capacity-exhaustion-fallback-lanes.html Sun, 29 Mar 2026 00:00:00 +0000 LiteLLM 1.82.7–1.82.8 were malicious PyPI releases tied to a compromised Trivy CI/CD path. This operator-focused write-up covers the fast audit, rotation, inspection, and pinning checklist for OpenClaw users. https://anyech.github.io/jingxiao-cai-blog/litellm-supply-chain-attack-openclaw.html https://anyech.github.io/jingxiao-cai-blog/litellm-supply-chain-attack-openclaw.html Sat, 28 Mar 2026 00:00:00 +0000 After proving local memory search worked, I stabilized a remote memory-only lane in OpenClaw. The follow-up reinforced the same lesson: source discipline, lexical anchors, and hybrid retrieval mattered more than another round of model churn. https://anyech.github.io/jingxiao-cai-blog/text-embedding-3-small-openclaw-remote-memory.html https://anyech.github.io/jingxiao-cai-blog/text-embedding-3-small-openclaw-remote-memory.html Fri, 27 Mar 2026 00:00:00 +0000 A debugging story about hidden platform caps, now updated with Anthropic's March 2026 flat 1M Claude pricing change and why cheaper long context still doesn't eliminate API-level ceilings. https://anyech.github.io/jingxiao-cai-blog/hidden-input-limit-llm-apis.html https://anyech.github.io/jingxiao-cai-blog/hidden-input-limit-llm-apis.html Tue, 24 Feb 2026 00:00:00 +0000 I run autonomous cron jobs with no built-in undo capability. When Moltbook's community started talking about recovery primitives, I realized that unattended automation needs a stronger recovery story. https://anyech.github.io/jingxiao-cai-blog/recovery-problem-ai-agent-undo.html https://anyech.github.io/jingxiao-cai-blog/recovery-problem-ai-agent-undo.html Sat, 07 Mar 2026 00:00:00 +0000 I created a sanitization checklist after nearly publishing sensitive deployment details. Here's what to redact, what to keep, and validation scripts for technical bloggers. https://anyech.github.io/jingxiao-cai-blog/blog-sanitization-checklist.html https://anyech.github.io/jingxiao-cai-blog/blog-sanitization-checklist.html Tue, 03 Mar 2026 00:00:00 +0000 A historical custom-skill loading bug, refreshed with a generated-Markdown regression lesson and a standalone artifact-contract follow-up. https://anyech.github.io/jingxiao-cai-blog/workspace-skills-bug.html https://anyech.github.io/jingxiao-cai-blog/workspace-skills-bug.html Mon, 23 Feb 2026 00:00:00 +0000 How I built an AI assistant to automate my morning email routine. https://anyech.github.io/jingxiao-cai-blog/personal-ai-assistant-gmail-morning-memo.html https://anyech.github.io/jingxiao-cai-blog/personal-ai-assistant-gmail-morning-memo.html Sat, 21 Feb 2026 00:00:00 +0000 Common issues and solutions when working with AI agent skills. https://anyech.github.io/jingxiao-cai-blog/troubleshooting-ai-agent-skills.html https://anyech.github.io/jingxiao-cai-blog/troubleshooting-ai-agent-skills.html Wed, 18 Feb 2026 00:00:00 +0000 How I migrated my cloud storage using MultCloud. https://anyech.github.io/jingxiao-cai-blog/onedrive-multcloud-story.html https://anyech.github.io/jingxiao-cai-blog/onedrive-multcloud-story.html Wed, 18 Feb 2026 00:00:00 +0000 A comprehensive guide to setting up Google APIs for your projects. https://anyech.github.io/jingxiao-cai-blog/google-api-setup-guide.html https://anyech.github.io/jingxiao-cai-blog/google-api-setup-guide.html Wed, 18 Feb 2026 00:00:00 +0000 After several years in industry, I'm starting to share my learnings, thoughts, and experiences... https://anyech.github.io/jingxiao-cai-blog/welcome.html https://anyech.github.io/jingxiao-cai-blog/welcome.html Tue, 17 Feb 2026 00:00:00 +0000 I used to run a blog on WordPress back in 2016-2017 during my PhD years at University of Oklahoma... https://anyech.github.io/jingxiao-cai-blog/migrating-from-wordpress.html https://anyech.github.io/jingxiao-cai-blog/migrating-from-wordpress.html Tue, 17 Feb 2026 00:00:00 +0000