Practice · 2025–now · AI Present
Agent skills / SKILL.md
Packaged workflows in a folder — name, description, scripts — loaded on demand instead of stuffing the system prompt.
Why Agent skills / SKILL.md stuck
Agent skills stuck because prompt engineering as a job title died and agents still needed procedures. SKILL.md (Anthropic, then an open format) plus AGENTS.md gave repos a portable way to teach tools without burning the context window on every turn. They fail when every README becomes a skill and nobody measures whether the agent follows it. The durable residue is treating workflows like code with progressive disclosure; the fad is a marketplace of magic folders.
Patterns
Compare with
Related
Practice · 2022–2025
Prompt engineering
Craft the magic string until the model behaves. Job title of 2023; table-stakes skill of 2026.
Practice · 2021–now
AI pair programming
Coding agents that edit the repo — now table stakes. Generation outran review capacity; ownership is the bottleneck.
Architecture · 2024–now
Context / tool protocols
Standard ways for models to talk to tools and data — MCP after the 2026 stateless rewrite, not one-off plugin snowflakes.
Platform · 2025–now
Vendor agent APIs
The Codex/Claude loop as a product SKU — durable sessions, compaction, subagents, and a sandbox you do not own.
$ Teams replaced a weekend of LangGraph glue with a managed session API, then discovered residency, recovery, and kill switches all live on the vendor side of the harness.
Practice · 2023–now
LLM eval pipelines
Regression tests for nondeterministic models that actually fail the build. The unglamorous CI that separates demos from products.