Content
72%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 well-architected overview skill: lean, high-signal warnings and troubleshooting tables, real one-level-deep reference files with executable content, and clear routing. The main weaknesses are the duplicated navigation tables and the absence of explicit validation checkpoints in the batch fleet workflows, which the rubric caps at 3 for workflow clarity.
Suggestions
Merge the 'Which do you need?' and 'Quick Navigation' tables into one routing table (or make one cover decisions and the other tasks without overlapping rows) to cut duplicated routing tokens.
Add explicit verification steps to the batch workflows, e.g. after instance refresh: check `aws autoscaling describe-instance-refresh` reaches Successful and instances pass ELB health checks before declaring the rollout done.
Include one minimal end-to-end command example for the flagship 'stand up an autoscaling web fleet' workflow (create-launch-template then create-auto-scaling-group) inline, deferring variants to references.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is dense and earns its tokens — specific facts like "default `HttpPutResponseHopLimit` of 1", "heartbeat 3600s, default ABANDON", and a compact symptom/cause/fix table with no padding or explanations of concepts Claude already knows. It is not a 5 because the "Which do you need?" table and the "Quick Navigation" table substantially overlap, routing to the same four reference files twice. | 4 / 5 |
Actionability | Concrete, executable specifics appear throughout: `--health-check-type ELB`, `HttpPutResponseHopLimit=2`, `AmazonSSMManagedInstanceCore`, `HttpTokens=required`, `price-capacity-optimized`, "Call `complete-lifecycle-action` CONTINUE", "Stop/start to migrate to new hardware". It is not a 5 because the full command sequences for the workflows (create launch template, create ASG, run instance refresh) are deferred to the references rather than shown, leaving minor gaps in copy-paste readiness at the top level. | 4 / 5 |
Workflow Clarity | The Common Workflows section sequences multi-step processes clearly ("Create a launch template (AMI, type, IMDSv2), then an ASG referencing it with `--health-check-type ELB` and a target-tracking policy"; "New launch template version → instance refresh; pin a numeric launch-template version so rollback works"), but no validation checkpoints are stated (e.g., verify instances pass ELB health checks before/after an instance refresh). Since fleet rollout via instance refresh is a batch operation, the rubric caps workflow clarity at 3 for missing validation/verification steps, despite the good sequencing. | 3 / 5 |
Progressive Disclosure | The SKILL.md is a genuine overview: two decision/navigation tables, four critical warnings each deep-linked to the correct file, workflow pointers, and a Not-Covered section. All six referenced files (auto-scaling.md, provisioning.md, instance-selection.md, systems-manager.md, ami-management.md, troubleshooting.md) exist, are one level deep, and cross-link only as siblings — matching the clear-overview, well-signaled-references anchor. Not score 4 because there are no real organization gaps beyond the table overlap already counted under conciseness. | 5 / 5 |
Total | 16 / 20 Passed |