What problem does it solve? Authoring a new skill for the Anakot Agent repository requires matching strict validator rules and peer conventions—frontmatter format, description length limits, directory placement, and section structure—and mistakes cause validation failures or skills that look inconsistent with the rest of the repo. ## Core Features & Use Cases - Validator-Compliant Frontmatter: Documents the exact requirements enforced by tools/skill_manager_tool.py::_validate_frontmatter, including the 1024-character description limit and 100,000-character file cap. - Peer-Matched Structure: Provides the standard section layout (Overview, When to Use, Common Pitfalls, Verification Checklist) used by every skill under skills/software-development/. - Correct Tooling Workflow: Explains when to use write_file versus skill_manage(action='create'), since the latter writes to ~/.anakot/skills/ instead of the repo tree. - Use Case: You need to commit a new reusable workflow skill to the anakot-agent repository on a feature branch, so you follow this guide to draft, validate locally with a Python snippet, and commit a conforming SKILL.md. ## Quick Start Ask the agent to create a new in-repo skill under skills/<category>/<name>/SKILL.md following the anakot-agent-skill-authoring conventions and validate it before committing.
