Blog
Notes, field reports and analysis on enterprise AI.
RSS feedAgent plugins bundle skills, MCP servers and hooks into one install for one client. A skilder role serves skills and scoped tools to any MCP agent from a governed workspace. How they differ, and when to use which.
Read article →A 30-page white paper tests whether serving skills through one MCP server can enforce what an agent may do. 13 scenarios, six models, three tool interfaces, and the limits the authors report.
Read article →LiteLLM fronts the models, agentgateway fronts the tools and agents, skilder defines the roles an agent performs. What each one governs, where it sits in the request path, and which one to deploy first.
Read article →Agents inherit whatever their tools can reach. Least privilege for AI agents means scoping tools per role, not per user, and treating every tool description as untrusted input.
Read article →Shadow MCP is shadow AI with hands: MCP servers employees connect to their AI clients outside any review. What it is, why it is riskier, and how to govern it without blocking.
Read article →Everyone is building agent skills, but almost no one can say which ones are worth keeping. Why skill value resists measurement, and what an honest answer looks like.
Read article →Skilder has been selected by the Venture Kick jury and awarded a CHF 10'000 Stage 1 grant.
Read article →There's a quiet architectural decision buried in most AI agents, and it shapes everything downstream: where does the agent's capability actually live?
Read article →Why Agent Skills, as currently architected, are becoming the single biggest accelerator of Shadow AI in the enterprise. And what to do about it.
Read article →Skilder has been awarded a CHF 20,000 Digital Grant by the Foundation for Innovation and Technology (FIT), the Vaud-based foundation that supports early-stage innovative companies in the canton.
Read article →Shadow AI is the unmanaged use of generative AI tools by employees. Definition, real-world risks, and how to respond without killing productivity.
Read article →95% of enterprise AI pilots fail not because models are weak, but because they're loaded with knowledge and asked to deliver know-how. Why these are different categories.
Read article →Why enterprise AI gets heavier as it gets more capable, and how a capability graph lets you assemble context as a runtime payload instead of a static config file.
Read article →A new layer is taking shape in the agentic enterprise stack: the context lake. Business context, tool permissions, and governance need their own home.
Read article →MCP solves the N×M connectivity problem. Skills solve context saturation. The real opportunity isn't picking one. It's composing skills over MCP, organized by business context.
Read article →Single skill files hit a complexity ceiling fast. Skill graphs turn domain expertise into a navigable network so agents can traverse business logic instead of guessing.
Read article →Workflow platforms vs. AI agent: it depends on how messy your reality is. A framework for choosing between deterministic workflows and adaptive agent skills, plus the hybrid that wins.
Read article →RAG retrieves content. Skills package competencies. Why conflating them limits how companies think about AI agents, and a framework for choosing between them.
Read article →As agents gain access to more tools, the bottleneck shifts from reasoning to retrieval. Why picking the right tool is the next frontier in AI agent design.
Read article →Agent Skills are modular instruction sets that extend AI capabilities for specific tasks. Anatomy of a SKILL.md, activation flow, and how to build your first one.
Read article →