Content
82%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 strong, command-driven workflow that covers the full plugin onboarding lifecycle with concrete, executable steps, an explicit test checkpoint, and well-signaled external references. Minor improvements are possible: reduce repeated location/command scaffolding, add a fix-and-retry loop after the integration test, and move the optional separate-workflow setup into a reference file.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is efficient and almost entirely repo-specific procedural knowledge Claude cannot know (scaffold commands, hooks fragments, telemetry patterns, pinned-SHA guidance), with no padding explaining known concepts. It sits at score 4 rather than 5 due to minor trimmable redundancy, e.g. the "cd scripts" blocks each duplicate the "# At <repo-root>/scripts/" location comments and the plugin-name instruction is repeated across sections. | 4 / 5 |
Actionability | Every step is a copy-paste-ready command with explicit working directories ("npm i", "npm run plugin:new -- --plugin {plugin-name}", "npm run build", "npm run test:vally -- --plugin {plugin-name} --skill {skill-name}"), including a failure path ("If it fails, ask the skill author to report a bug to microsoft/github-copilot-for-azure"). This matches score 5's fully executable, common-case-covering anchor; score 4 would require minor gaps in the commands, which are absent. | 5 / 5 |
Workflow Clarity | The sequence is clear and well-ordered (install deps → scaffold plugin → scaffold skills → build → run integration test → prepare release → finish implementation) with an explicit checkpoint: "The test should pass. If it fails, ask the skill author to report a bug." It stops short of score 5 because the failure path defers to bug reporting rather than a validate-fix-retry loop, and there is no verification that scaffolded files were created before proceeding to the test step. | 4 / 5 |
Progressive Disclosure | Sections are well organized and external material is appropriately deferred via clearly signaled, one-level-deep references ("Refer to the `Plugin Structure` section of [Onboarding.md]", "Follow [skill-authoring]", "Follow [vally-eval]") — matching score 4's good structure with minor gaps. It is not score 5 because the ~25-line "Set up a separate integration test workflow" optional section is inlined detail that would fit better in a reference file. | 4 / 5 |
Total | 17 / 20 Passed |