<?xml version='1.0' encoding='UTF-8'?>
<rss xmlns:atom="http://www.w3.org/2005/Atom" version="2.0">
  <channel>
    <title>Ghassan Alhamoud — AI Architecture &amp; Automation</title>
    <link>https://ghassan-alhamoud.com</link>
    <description>AI architecture, systems engineering, distributed systems, and the hard problems nobody’s solving — by Ghassan Alhamoud.</description>
    <language>en</language>
    <lastBuildDate>Mon, 27 Jul 2026 09:42:39 GMT</lastBuildDate>
    <atom:link href="https://ghassan-alhamoud.com/rss.xml" rel="self" type="application/rss+xml" />
    <item>
      <title>I Stopped Guessing What to Automate. So I Built an Open-Source Miner.</title>
      <link>https://ghassan-alhamoud.com/articles/automation-miner-open-source.html</link>
      <guid isPermaLink="true">https://ghassan-alhamoud.com/articles/automation-miner-open-source.html</guid>
      <pubDate>Sun, 26 Jul 2026 22:00:00 GMT</pubDate>
      <description><![CDATA[Most AI projects start with a solution and search for a problem. Automation Miner turns domain evidence into a citable, scored portfolio of opportunities worth investigating.]]></description>
      <category>Open Source</category>
      <category>Automation</category>
      <category>AI Agents</category>
      <category>LangGraph</category>
      <category>MCP</category>
      <category>Python</category>
    </item>
    <item>
      <title>Your Agent Does Not Need Inbox Access. It Needs Email Intelligence.</title>
      <link>https://ghassan-alhamoud.com/articles/email-intelligence-protocol.html</link>
      <guid isPermaLink="true">https://ghassan-alhamoud.com/articles/email-intelligence-protocol.html</guid>
      <pubDate>Thu, 23 Jul 2026 22:00:00 GMT</pubDate>
      <description><![CDATA[Raw mailbox access gave my agents messages, not understanding — and exposed them to untrusted content. So I built an open-source, local-first email intelligence layer.]]></description>
      <category>Open Source</category>
      <category>AI Agents</category>
      <category>MCP</category>
      <category>Email</category>
      <category>Security</category>
      <category>Local First</category>
    </item>
    <item>
      <title>Testing Terminal UIs with tmux: A Deterministic PTY Workflow</title>
      <link>https://ghassan-alhamoud.com/articles/testing-tuis-with-tmux.html</link>
      <guid isPermaLink="true">https://ghassan-alhamoud.com/articles/testing-tuis-with-tmux.html</guid>
      <pubDate>Wed, 01 Jul 2026 22:00:00 GMT</pubDate>
      <description><![CDATA[A copyable tmux workflow for black-box TUI tests: isolated PTY, explicit dimensions, state polling, focus handling, terminal-size matrix, CI boundary, and fixture script.]]></description>
      <category>TUI</category>
      <category>Testing</category>
      <category>tmux</category>
      <category>Python</category>
      <category>Automation</category>
      <category>DevOps</category>
    </item>
    <item>
      <title>From Agent Experience to Portable Knowledge: Integrating ALMS and OKF</title>
      <link>https://ghassan-alhamoud.com/articles/alms-okf-integration-bridge.html</link>
      <guid isPermaLink="true">https://ghassan-alhamoud.com/articles/alms-okf-integration-bridge.html</guid>
      <pubDate>Wed, 01 Jul 2026 22:00:00 GMT</pubDate>
      <description><![CDATA[A governed promotion path from ALMS operational learnings to portable OKF bundles: thresholds, provenance, conflicts, security, and version compatibility.]]></description>
      <category>AI Agents</category>
      <category>OKF</category>
      <category>ALMS</category>
      <category>Knowledge Management</category>
      <category>MCP</category>
      <category>Agent Memory</category>
    </item>
    <item>
      <title>Why a Strong Coding Agent Failed a TUI Task: A Detailed Case Study</title>
      <link>https://ghassan-alhamoud.com/articles/smart-models-go-stupid.html</link>
      <guid isPermaLink="true">https://ghassan-alhamoud.com/articles/smart-models-go-stupid.html</guid>
      <pubDate>Tue, 23 Jun 2026 22:00:00 GMT</pubDate>
      <description><![CDATA[A repository-backed case study: 6,142 lines of TUI design became a 280-line input/print REPL. Here is the acceptance gap and the workflow that later produced the real application.]]></description>
      <category>AI Agents</category>
      <category>Code Generation</category>
      <category>Engineering</category>
      <category>UX</category>
      <category>LLM Limitations</category>
      <category>TUI</category>
      <category>Terminal</category>
    </item>
    <item>
      <title>SkillLedger: A Verification and Entitlement Layer for Agent Skills</title>
      <link>https://ghassan-alhamoud.com/articles/skill-ledger-open-source-verified-agent-skill-marketplace.html</link>
      <guid isPermaLink="true">https://ghassan-alhamoud.com/articles/skill-ledger-open-source-verified-agent-skill-marketplace.html</guid>
      <pubDate>Sun, 14 Jun 2026 22:00:00 GMT</pubDate>
      <description><![CDATA[Inside SkillLedger's MVP trust model: versioned manifests, deterministic integrity checks, separate approval, entitlements, ledger entries, and explicit limits.]]></description>
      <category>Open Source</category>
      <category>Agents</category>
      <category>MCP</category>
      <category>Ruby on Rails</category>
      <category>Marketplace</category>
      <category>API</category>
      <category>Infrastructure</category>
      <category>Agent Skills</category>
    </item>
    <item>
      <title>ALMS: Shared Operational Memory for Autonomous Agents via MCP</title>
      <link>https://ghassan-alhamoud.com/articles/alms-open-source-shared-agent-memory.html</link>
      <guid isPermaLink="true">https://ghassan-alhamoud.com/articles/alms-open-source-shared-agent-memory.html</guid>
      <pubDate>Wed, 10 Jun 2026 22:00:00 GMT</pubDate>
      <description><![CDATA[Inside ALMS 0.1.0: a Go/PostgreSQL MCP control plane for shared operational learnings, gap-safe sync, protocols, OKF export, and explicit limits.]]></description>
      <category>Open Source</category>
      <category>Agents</category>
      <category>MCP</category>
      <category>Go</category>
      <category>PostgreSQL</category>
      <category>Multi-Agent</category>
      <category>Learning</category>
      <category>Infrastructure</category>
    </item>
    <item>
      <title>Quality vs Speed with AI Agents: A Risk-Based Gate Framework</title>
      <link>https://ghassan-alhamoud.com/articles/quality-vs-speed-agent-era.html</link>
      <guid isPermaLink="true">https://ghassan-alhamoud.com/articles/quality-vs-speed-agent-era.html</guid>
      <pubDate>Wed, 03 Jun 2026 22:00:00 GMT</pubDate>
      <description><![CDATA[A practical framework for routing AI-agent work through fast, standard, or critical quality gates based on impact, reversibility, sensitivity, authority, and verifiability.]]></description>
      <category>AI Agents</category>
      <category>Quality Gates</category>
      <category>Engineering</category>
      <category>Risk Management</category>
    </item>
    <item>
      <title>Building an AI Film Pipeline: Architecture, Failures, and Current Evidence</title>
      <link>https://ghassan-alhamoud.com/articles/film-pipeline-postmortem.html</link>
      <guid isPermaLink="true">https://ghassan-alhamoud.com/articles/film-pipeline-postmortem.html</guid>
      <pubDate>Mon, 01 Jun 2026 22:00:00 GMT</pubDate>
      <description><![CDATA[An evidence-backed postmortem covering historical production failures, eight-phase architecture, current scope, and a 1,840-test verification snapshot.]]></description>
      <category>AI Film</category>
      <category>Pipeline</category>
      <category>Postmortem</category>
      <category>Seedance</category>
      <category>Veo</category>
      <category>Engineering</category>
    </item>
    <item>
      <title>Auditing a 12-Hour AI Go Code-Generation Marathon</title>
      <link>https://ghassan-alhamoud.com/articles/ai-coded-83-mcp-tools-marathon.html</link>
      <guid isPermaLink="true">https://ghassan-alhamoud.com/articles/ai-coded-83-mcp-tools-marathon.html</guid>
      <pubDate>Sat, 30 May 2026 22:00:00 GMT</pubDate>
      <description><![CDATA[A repository-backed audit of a 12-hour AI-assisted Go marathon: 32 measured feature commits, tool growth, a 97-parameter constructor, and a reusable release gate.]]></description>
      <category>AI Agents</category>
      <category>Go</category>
      <category>Architecture</category>
      <category>MCP</category>
      <category>Automation</category>
      <category>Code Generation</category>
      <category>Engineering</category>
    </item>
    <item>
      <title>Agent-First Entity Database: Schema, MCP Contract, and Benchmarks</title>
      <link>https://ghassan-alhamoud.com/articles/entity-database-zero-ui-agents.html</link>
      <guid isPermaLink="true">https://ghassan-alhamoud.com/articles/entity-database-zero-ui-agents.html</guid>
      <pubDate>Wed, 27 May 2026 22:00:00 GMT</pubDate>
      <description><![CDATA[Inside EntityScope: a zero-UI DuckDB knowledge base for agents, with field-level provenance, bounded MCP tools, measured query latency, test results, and reusable contracts.]]></description>
      <category>AI Agents</category>
      <category>MCP</category>
      <category>DuckDB</category>
      <category>Data Architecture</category>
    </item>
    <item>
      <title>Structured Agent Handoffs: Contracts, State Machines, and Acceptance Gates</title>
      <link>https://ghassan-alhamoud.com/articles/orchestrator-qwen-acp-handoff.html</link>
      <guid isPermaLink="true">https://ghassan-alhamoud.com/articles/orchestrator-qwen-acp-handoff.html</guid>
      <pubDate>Wed, 27 May 2026 22:00:00 GMT</pubDate>
      <description><![CDATA[A production pattern for delegating coding work between agents: typed scope, permissions, idempotency, lifecycle states, retries, acceptance evidence, and hard-won integration failures.]]></description>
      <category>AI Agents</category>
      <category>Delegation</category>
      <category>MCP</category>
      <category>Reliability</category>
    </item>
    <item>
      <title>How I Built a 3 ms Workspace Index for AI Agents</title>
      <link>https://ghassan-alhamoud.com/articles/workspace-indexing-3ms.html</link>
      <guid isPermaLink="true">https://ghassan-alhamoud.com/articles/workspace-indexing-3ms.html</guid>
      <pubDate>Tue, 05 May 2026 22:00:00 GMT</pubDate>
      <description><![CDATA[A reproducible benchmark of a 302 KB semantic registry over 2,408 files, plus its schema, update strategy, correctness gates, and retrieval tradeoffs.]]></description>
      <category>Architecture</category>
      <category>Performance</category>
      <category>Workspace Indexing</category>
      <category>AI</category>
      <category>OpenClaw</category>
    </item>
    <item>
      <title>Self-Healing Agent Systems: A Bounded Remediation State Machine</title>
      <link>https://ghassan-alhamoud.com/articles/self-healing-architecture-feedback-loop.html</link>
      <guid isPermaLink="true">https://ghassan-alhamoud.com/articles/self-healing-architecture-feedback-loop.html</guid>
      <pubDate>Mon, 04 May 2026 22:00:00 GMT</pubDate>
      <description><![CDATA[A production design for self-healing agent runtimes: typed detection, confidence, preconditions, idempotency, bounded remediation, verification, compensation, circuit breakers, and escalation.]]></description>
      <category>Self-Healing</category>
      <category>Reliability</category>
      <category>State Machines</category>
      <category>Circuit Breakers</category>
    </item>
    <item>
      <title>Issue Management as a Reliability Control Plane</title>
      <link>https://ghassan-alhamoud.com/articles/issue-management-system-property.html</link>
      <guid isPermaLink="true">https://ghassan-alhamoud.com/articles/issue-management-system-property.html</guid>
      <pubDate>Sun, 03 May 2026 22:00:00 GMT</pubDate>
      <description><![CDATA[A durable issue-control-plane design for agent systems: fingerprints, lifecycle states, evidence-backed root cause, corrective-action classes, verification windows, recurrence, and machine-readable records.]]></description>
      <category>Reliability</category>
      <category>Incident Management</category>
      <category>Root Cause</category>
      <category>Evidence</category>
    </item>
    <item>
      <title>Five Failures From Three Months Building a Personal AI Agent</title>
      <link>https://ghassan-alhamoud.com/articles/5-failures-building-agent-3-months.html</link>
      <guid isPermaLink="true">https://ghassan-alhamoud.com/articles/5-failures-building-agent-3-months.html</guid>
      <pubDate>Sun, 03 May 2026 22:00:00 GMT</pubDate>
      <description><![CDATA[A documented failure timeline from three months operating a personal AI agent: provider loss, daemon outages, unsafe self-modification, reboot fragility, and unverified fixes.]]></description>
      <category>AI Agents</category>
      <category>Infrastructure</category>
      <category>Failure</category>
      <category>Architecture</category>
      <category>AI</category>
      <category>OpenClaw</category>
    </item>
    <item>
      <title>Five Infrastructure Layers Every Production AI Agent Needs</title>
      <link>https://ghassan-alhamoud.com/articles/5-infrastructure-layers-broke-agent.html</link>
      <guid isPermaLink="true">https://ghassan-alhamoud.com/articles/5-infrastructure-layers-broke-agent.html</guid>
      <pubDate>Sun, 03 May 2026 22:00:00 GMT</pubDate>
      <description><![CDATA[A production AI agent reference architecture covering transport, policy, orchestration, verification and recovery, and observability—with interfaces, failure modes, and readiness gates.]]></description>
      <category>AI Agents</category>
      <category>Infrastructure</category>
      <category>Architecture</category>
      <category>Reliability</category>
      <category>Security</category>
    </item>
    <item>
      <title>What a 250-Item Agent Run Actually Proved</title>
      <link>https://ghassan-alhamoud.com/articles/stress-test-self-healing-250-target.html</link>
      <guid isPermaLink="true">https://ghassan-alhamoud.com/articles/stress-test-self-healing-250-target.html</guid>
      <pubDate>Sat, 02 May 2026 22:00:00 GMT</pubDate>
      <description><![CDATA[An evidence-backed postmortem of a 252-item workflow, plus a reproducible 250-case protocol that tests recovery without confusing throughput with self-healing.]]></description>
      <category>Reliability</category>
      <category>Evaluation</category>
      <category>Fault Injection</category>
      <category>AI Agents</category>
      <category>OpenClaw</category>
    </item>
    <item>
      <title>The Same Automation Pattern Across Six Industries: A Comparative Study</title>
      <link>https://ghassan-alhamoud.com/articles/automation-pattern-6-industries.html</link>
      <guid isPermaLink="true">https://ghassan-alhamoud.com/articles/automation-pattern-6-industries.html</guid>
      <pubDate>Fri, 01 May 2026 22:00:00 GMT</pubDate>
      <description><![CDATA[A six-domain comparison of top-scoring automation candidates: triggers, bottlenecks, available data, risks, expected value, confidence, and selection bias.]]></description>
      <category>Automation</category>
      <category>Discovery</category>
      <category>Architecture</category>
      <category>Pattern</category>
      <category>AI</category>
      <category>OpenClaw</category>
    </item>
    <item>
      <title>How an AI System Found and Scored 42 Automation Opportunities</title>
      <link>https://ghassan-alhamoud.com/articles/ai-found-42-automation-opportunities.html</link>
      <guid isPermaLink="true">https://ghassan-alhamoud.com/articles/ai-found-42-automation-opportunities.html</guid>
      <pubDate>Fri, 01 May 2026 22:00:00 GMT</pubDate>
      <description><![CDATA[The published 42-item dataset behind a six-domain automation-discovery experiment, including score distribution, method, selection bias, and validation gates.]]></description>
      <category>Automation</category>
      <category>Discovery</category>
      <category>Dataset</category>
      <category>Scoring</category>
      <category>Validation</category>
      <category>OpenClaw</category>
    </item>
    <item>
      <title>Safe Self-Improvement Automation: Detect, Repair, Verify, Roll Back</title>
      <link>https://ghassan-alhamoud.com/articles/most-boring-automation-self-improving.html</link>
      <guid isPermaLink="true">https://ghassan-alhamoud.com/articles/most-boring-automation-self-improving.html</guid>
      <pubDate>Tue, 14 Apr 2026 22:00:00 GMT</pubDate>
      <description><![CDATA[A production pattern for bounded self-improvement automation: declarative inventory, evidence-backed findings, risk-classed auto-fixes, preconditions, verification, rollback, reconciliation, and honest metrics.]]></description>
      <category>Automation</category>
      <category>Feedback Loops</category>
      <category>Verification</category>
      <category>Rollback</category>
    </item>
    <item>
      <title>Agent-to-Agent Negotiation by Email: A Pre-Experiment Protocol</title>
      <link>https://ghassan-alhamoud.com/articles/a2a-negotiation-outreach.html</link>
      <guid isPermaLink="true">https://ghassan-alhamoud.com/articles/a2a-negotiation-outreach.html</guid>
      <pubDate>Sat, 11 Apr 2026 22:00:00 GMT</pubDate>
      <description><![CDATA[An honest pre-experiment report for agent negotiation over email: what was built, what was not tested, the required metrics, safety controls, and a reproducible protocol.]]></description>
      <category>AI Agents</category>
      <category>Negotiation</category>
      <category>Experimental Design</category>
      <category>Email</category>
    </item>
    <item>
      <title>How I Use an LLM to Prepare for a High-Stakes Negotiation</title>
      <link>https://ghassan-alhamoud.com/articles/negotiation-veteran-collapse-llm-coach.html</link>
      <guid isPermaLink="true">https://ghassan-alhamoud.com/articles/negotiation-veteran-collapse-llm-coach.html</guid>
      <pubDate>Thu, 09 Apr 2026 22:00:00 GMT</pubDate>
      <description><![CDATA[A privacy-conscious LLM negotiation-preparation protocol: define decision boundaries, rehearse difficult scenarios, challenge assumptions, and debrief without outsourcing judgment.]]></description>
      <category>Negotiation</category>
      <category>LLM</category>
      <category>Preparation</category>
      <category>Decision Support</category>
    </item>
    <item>
      <title>Designing a 90% Automated Product Pipeline—What Must Stay Human</title>
      <link>https://ghassan-alhamoud.com/articles/dreamforge-automation-pipeline.html</link>
      <guid isPermaLink="true">https://ghassan-alhamoud.com/articles/dreamforge-automation-pipeline.html</guid>
      <pubDate>Tue, 07 Apr 2026 22:00:00 GMT</pubDate>
      <description><![CDATA[An honest architecture case study of DreamForge: what agents can own from idea to pull request, where human approval remains essential, and how to validate automation safely.]]></description>
      <category>Automation</category>
      <category>Product Delivery</category>
      <category>AI Agents</category>
      <category>Human-in-the-Loop</category>
    </item>
    <item>
      <title>Designing an On-Device Personal Memory System</title>
      <link>https://ghassan-alhamoud.com/articles/echoweave-on-device-narrative-os.html</link>
      <guid isPermaLink="true">https://ghassan-alhamoud.com/articles/echoweave-on-device-narrative-os.html</guid>
      <pubDate>Sat, 04 Apr 2026 22:00:00 GMT</pubDate>
      <description><![CDATA[EchoWeave is an unbuilt design proposal for private personal memory: an on-device architecture, consent and deletion model, threat analysis, resource constraints, and validation plan.]]></description>
      <category>Privacy</category>
      <category>On-Device AI</category>
      <category>Personal Memory</category>
      <category>Threat Modeling</category>
    </item>
    <item>
      <title>Agent Memory Architecture: Admission, Retrieval, Consolidation, and Forgetting</title>
      <link>https://ghassan-alhamoud.com/articles/every-token-is-rent-memory-architecture.html</link>
      <guid isPermaLink="true">https://ghassan-alhamoud.com/articles/every-token-is-rent-memory-architecture.html</guid>
      <pubDate>Tue, 31 Mar 2026 22:00:00 GMT</pubDate>
      <description><![CDATA[A production lifecycle for agent memory: memory classes, admission gates, provenance-aware retrieval, consolidation, contradiction handling, decay, deletion, evaluation, and cost accounting.]]></description>
      <category>Agent Memory</category>
      <category>Retrieval</category>
      <category>Knowledge Systems</category>
      <category>Evaluation</category>
    </item>
    <item>
      <title>StartupGraph in Production: A Self-Hosted Startup Intelligence System</title>
      <link>https://ghassan-alhamoud.com/articles/startupgraph-insights-engine.html</link>
      <guid isPermaLink="true">https://ghassan-alhamoud.com/articles/startupgraph-insights-engine.html</guid>
      <pubDate>Sun, 29 Mar 2026 22:00:00 GMT</pubDate>
      <description><![CDATA[A production-minded case study of StartupGraph: 1,010 entities, field-level provenance, agent queries over MCP, and the limits of a self-hosted startup intelligence system.]]></description>
      <category>Startup Intelligence</category>
      <category>MCP</category>
      <category>DuckDB</category>
      <category>Data Engineering</category>
    </item>
    <item>
      <title>AI Agent Execution Attestation: A Verifiable Audit Trail</title>
      <link>https://ghassan-alhamoud.com/articles/agent-execution-attestation.html</link>
      <guid isPermaLink="true">https://ghassan-alhamoud.com/articles/agent-execution-attestation.html</guid>
      <pubDate>Fri, 27 Mar 2026 23:00:00 GMT</pubDate>
      <description><![CDATA[A vendor-neutral execution-attestation design for AI agents: envelope schema, threat model, signing and verification flow, effect receipts, and a cryptographically verifiable example.]]></description>
      <category>AI Agents</category>
      <category>Attestation</category>
      <category>Auditability</category>
      <category>Security</category>
      <category>Cryptography</category>
    </item>
    <item>
      <title>Compliance as Code for the EU AI Act: An Engineering Framework</title>
      <link>https://ghassan-alhamoud.com/articles/compliance-as-code-eu-ai-act.html</link>
      <guid isPermaLink="true">https://ghassan-alhamoud.com/articles/compliance-as-code-eu-ai-act.html</guid>
      <pubDate>Tue, 24 Mar 2026 23:00:00 GMT</pubDate>
      <description><![CDATA[A source-linked engineering framework for EU AI Act evidence: applicability, versioned controls, CI change impact, runtime records, approval gates, and audit packages.]]></description>
      <category>EU AI Act</category>
      <category>Compliance Engineering</category>
      <category>AI Governance</category>
      <category>Auditability</category>
    </item>
    <item>
      <title>Agent Reliability SLOs: Measure Correctness, Safety, and Recovery</title>
      <link>https://ghassan-alhamoud.com/articles/agent-reliability-sla-infrastructure.html</link>
      <guid isPermaLink="true">https://ghassan-alhamoud.com/articles/agent-reliability-sla-infrastructure.html</guid>
      <pubDate>Sat, 21 Mar 2026 23:00:00 GMT</pubDate>
      <description><![CDATA[A practical SLI/SLO framework for production AI agents: eligible journeys, verified task success, unsafe and duplicate effects, evidence completeness, recovery time, error budgets, and burn-rate alerts.]]></description>
      <category>AI Agents</category>
      <category>Reliability</category>
      <category>SRE</category>
      <category>SLOs</category>
      <category>Observability</category>
    </item>
    <item>
      <title>Build an Ideation Engine: A Repeatable Product Discovery System</title>
      <link>https://ghassan-alhamoud.com/articles/systematic-ideation-engine.html</link>
      <guid isPermaLink="true">https://ghassan-alhamoud.com/articles/systematic-ideation-engine.html</guid>
      <pubDate>Thu, 19 Mar 2026 23:00:00 GMT</pubDate>
      <description><![CDATA[A practical product ideation system with three analytical lenses, hard quality gates, deduplication, a weighted scorecard, rejected ideas, and a copyable validation workflow.]]></description>
      <category>Ideation</category>
      <category>Product Discovery</category>
      <category>Validation</category>
      <category>AI Systems</category>
    </item>
    <item>
      <title>Build vs Buy Startup Intelligence: A Self-Hosted Entity Graph</title>
      <link>https://ghassan-alhamoud.com/articles/startup-intelligence-entity-graph.html</link>
      <guid isPermaLink="true">https://ghassan-alhamoud.com/articles/startup-intelligence-entity-graph.html</guid>
      <pubDate>Tue, 17 Mar 2026 23:00:00 GMT</pubDate>
      <description><![CDATA[A practical build-vs-buy framework grounded in a working DuckDB entity graph with field-level provenance, agent queries, and measured limitations.]]></description>
      <category>Startup Intelligence</category>
      <category>Entity Resolution</category>
      <category>Data Provenance</category>
      <category>DuckDB</category>
      <category>AI Agents</category>
    </item>
    <item>
      <title>The Security Risks of Unverified AI Agent Skills</title>
      <link>https://ghassan-alhamoud.com/articles/agent-black-box-skill-ledger.html</link>
      <guid isPermaLink="true">https://ghassan-alhamoud.com/articles/agent-black-box-skill-ledger.html</guid>
      <pubDate>Wed, 11 Mar 2026 23:00:00 GMT</pubDate>
      <description><![CDATA[A practical threat model for third-party agent skills: what manifests, checksums, approvals, signatures, sandboxes, and execution attestations can—and cannot—prove.]]></description>
      <category>AI Agents</category>
      <category>Agent Skills</category>
      <category>Security</category>
      <category>Software Supply Chain</category>
      <category>Threat Modeling</category>
    </item>
    <item>
      <title>Stop Building Agent Soup: Use Compiler-Style Agent Workflows</title>
      <link>https://ghassan-alhamoud.com/articles/stop-building-agent-soup-compilers.html</link>
      <guid isPermaLink="true">https://ghassan-alhamoud.com/articles/stop-building-agent-soup-compilers.html</guid>
      <pubDate>Mon, 09 Mar 2026 23:00:00 GMT</pubDate>
      <description><![CDATA[A concrete compiler-style pattern for agent workflows: normalize intent into an intermediate representation, validate capabilities and policy, execute deterministically, and verify effects.]]></description>
      <category>AI Agents</category>
      <category>Architecture</category>
      <category>Engineering</category>
      <category>OpenClaw</category>
    </item>
    <item>
      <title>AI Agent Discovery Is a Bigger Bottleneck Than Reasoning</title>
      <link>https://ghassan-alhamoud.com/articles/agent-reasoning-discovery-fallacy.html</link>
      <guid isPermaLink="true">https://ghassan-alhamoud.com/articles/agent-reasoning-discovery-fallacy.html</guid>
      <pubDate>Sat, 07 Mar 2026 23:00:00 GMT</pubDate>
      <description><![CDATA[A five-stage capability-discovery model for AI agents: registry, semantic matching, compatibility, trust qualification, and policy-aware runtime selection.]]></description>
      <category>AI Agents</category>
      <category>Discovery</category>
      <category>Distributed Systems</category>
      <category>AI</category>
      <category>OpenClaw</category>
    </item>
  </channel>
</rss>
