PostHog Engineering: How to Write Effective AI Agent Skills
rseroter · x · 2026-08-07
PostHog shares engineering insights on building and using AI agent skills internally. Their teams have published 226 skills to their internal skill store and maintain 187 SKILL.md files across their codebase.
The article highlights that the core pain point of agents is "amnesia"—rediscovering codebases and workflows from scratch in every conversation wastes time and tokens. Writing skills solves this, with key principles including:
- Progressive disclosure: Instead of dumping all context upfront, skills act as a router. The agent loads information on demand based on the skill's name and description.
- Be precise on goals: Skill definitions must clearly describe the goal, constraints, and non-derived context.
- Be ambiguous on steps: Leave execution steps, failure handling, and runtime specifics ambiguous to allow the agent to make its own decisions.
More from coding & agent
- Cua Driver Launches Extension-Free Browser Use Interface for AI Agents — rudrank · 2026-08-07
- LangChain Releases Guide on Open Source Agent Frameworks — LangChain · 2026-08-07
- Naïve Raises $28.5M Series A to Build Business Infrastructure for AI Agents — Scobleizer · 2026-08-07
- AI Coding Practice: Solid Planning Does 80% of the Job — EXM7777 · 2026-08-07
- HuggingFace CEO: Public Agent Collaboration Makes AI Safer and More Efficient — ben_burtenshaw · 2026-08-07
- Don't Let Coding Agents Grade Their Own Homework: Introducing QA Agents — sergeykarayev · 2026-08-07