Content
88%Weight 40%Scale 1-5Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
A lean, highly actionable skill: executable gh/GraphQL commands, a clear six-step workflow with validation feedback loops, approval gates on destructive actions, and a loop-termination rule. Minor room to grow in filling command placeholders and potentially splitting edge-case operations into a reference file.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is commands and decision rules only — no re-teaching of git/CI concepts, and the single explanatory line ("Unresolved threads block merging even after approval") earns its tokens. | 5 / 5 |
Actionability | Step 1's assessment commands and the GraphQL thread-resolution mutation are complete and executable, but recurring placeholders (<COMMENT_ID>, [commit], --body "...", git add <files>) leave minor gaps in copy-paste readiness. | 4 / 5 |
Workflow Clarity | Steps 1–6 are clearly sequenced with an explicit re-check loop ("Repeat Steps 1–5 until..." plus gh pr checks --watch), an escalation cap ("after 3 attempts, report the blocker"), and required user approval before destructive operations (force-push, re-create, delete). | 5 / 5 |
Progressive Disclosure | Well-organized numbered sections in a single self-contained file with no nested references, but at ~120 lines it exceeds the under-50-line simple-skill carve-out, and the Special Operations / GraphQL detail could plausibly live in a reference file. | 4 / 5 |
Total | 18 / 20 Passed |