When Agent Skills Lose to Simpler Guidance
A local Sigil benchmark found that, for summary and validation work, a small always-on guidance layer plus MCP outperformed manual invocation of a dedicated agent skill.
Research & Analysis
Analytical reports, operational frameworks, and research notes from WBA's independent practice.
48 Days of AI Memory: What the Productivity Data Actually Shows
Most organizations measure AI adoption by counting licenses, API calls, or chatbot sessions. These are activity metrics. They tell you tools are being used. They do not tell you whether…
Multi-Model AI Memory: What the Implementation Record Reveals
Most organizations deploying multiple AI models discover that each model operates in isolation — context evaporates between sessions, decisions are repeated, and institutional knowledge fails to accumulate. This analysis examines why multi-model memory is an organizational infrastructure problem and what the implementation record reveals about durable solutions.
The Cognitive Infrastructure Gap: Why AI Adoption Doesn’t Equal AI Maturity
Based on WBA market observations, over 70% of organizational AI activity remains trapped in conversational interfaces. This analysis introduces a three-stage maturity framework, a self-assessment diagnostic, and an economic model for understanding the gap between AI adoption and AI orchestration.
AI Productivity Is Not the Same as AI Understanding
AI tools are getting faster — but speed and understanding are not the same thing. When systems skip available context and rely on heuristics, the result is an illusion of memory. The next phase of AI maturity will be defined by epistemic discipline, not more automation.
From Epistemic Debt to Epistemic Instrumentation: MCP as a Business Pattern
Organizations often keep sounding confident long after they stop knowing. MCP — implemented with safety tiers, structured errors, and audit-grade logging — can be a practical step toward epistemic instrumentation.