Skills in this directory are for contributors working on the microsoft/winappcli
repository itself. They are read by Copilot CLI (and other agents) to perform
repo-specific developer tasks like reviewing a PR before push.
Not the same as
.github/plugin/skills/. That directory contains the shippedwinappCopilot plugin skills, which help end users of thewinappCLI tool. Those are auto-generated fromdocs/fragments/skills/viascripts/generate-llm-docs.ps1. Skills under.github/skills/are hand-written and not shipped to end users.
| Skill | Purpose |
|---|---|
pr-review/ |
Multi-dimensional review of a PR / feature branch diff (security, correctness, CLI UX, alternative solutions, tests, docs/samples, packaging, multi-model cross-check). Reports findings to stdout; does not apply fixes. |
spec-review/ |
Pre-code, decision-oriented review of a spec or proposed feature against the real codebase (necessity & scope, approach & alternatives, feasibility vs reality, risks/unknowns/edge cases, DX & user impact, multi-model cross-check). Reports a proceed / proceed-with-changes / reconsider recommendation to stdout; does not change code or the spec. |
- Each skill is a directory containing a
SKILL.md(the entry point the orchestrating agent reads) and any supporting prompt fragments. - Skills do not run scripts. The orchestrating agent uses its own tools
(
task,grep,view,powershellfor git, etc.) following the instructions inSKILL.md. - Prompt fragments meant to be passed verbatim to sub-agents live under a
dimensions/(or similarly named) subfolder. - Output goes to stdout unless the user explicitly asks for a file.