The description names the domain (GTM automation) and mentions some actions like 'build GTM automation with code,' 'design workflow architectures,' 'use AI agents for GTM tasks,' but these are somewhat abstract rather than concrete, specific actions. Phrases like 'implement the architecture over tools principle' and 'technical GTM infrastructure from workflow design through agent orchestration' are more conceptual than actionable.

The description clearly answers both 'what' (build GTM automation, design workflow architectures, use AI agents for GTM tasks, implement architecture over tools principle) and 'when' with explicit trigger guidance ('When the user wants to...' and 'Also use when the user mentions...'). It also includes negative boundaries ('Do NOT use for...').

Excellent coverage of natural trigger terms users would say: 'GTM engineering,' 'GTM automation,' 'n8n,' 'Make,' 'Zapier,' 'workflow automation,' 'Clay API,' 'instruction stacks,' 'AI agents for GTM,' 'revenue automation.' These are specific tool names and domain terms that users would naturally mention.

The description carves out a niche with GTM-specific terms and tool names, and the 'Do NOT use for technical implementation, code review, or software architecture' exclusion helps. However, there's potential overlap with general workflow automation skills or coding skills, and the contradiction between 'build GTM automation with code' and 'Do NOT use for technical implementation' creates confusion about boundaries.

Extremely verbose. Extensive sections explaining what GTM engineers do, career trajectories, role comparisons vs adjacent roles, and detailed platform comparison tables are information Claude already knows or can infer. The instruction stack concept explanation is padded with obvious details. The content is ~400+ lines when it could be under 100 with references.

Provides structured frameworks (instruction stack layers, decision trees, platform comparisons) and some concrete patterns (persistent context schema, feedback loop table), but lacks executable code examples, specific API calls, or copy-paste-ready workflow configurations. The examples section at the bottom describes outcomes rather than providing actual implementation steps.

The 'Before Starting' discovery checklist is well-structured, and the instruction stack layers provide a clear conceptual sequence. However, there are no explicit validation checkpoints, no feedback loops for error recovery during the build process itself, and the troubleshooting section is disconnected from a step-by-step workflow. The platform selection decision tree is helpful but doesn't constitute a build workflow.

References to 'references/implementation-guide.md' and 'references/quick-reference.md' show some progressive disclosure, and the related skills table is well-organized. However, the main file contains massive amounts of content that should be in reference files (the entire platform comparison section, the role comparison tables, the architecture vs tools framework). The overview itself is far too heavy.