Content
53%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.
The body is information-dense and largely actionable, with real Terraform and concrete savings figures, but it is essentially a monolithic reference dumped into SKILL.md: validation is implicit, generic boilerplate pads the opening, and every cross-reference points to files (and one wrong directory) that do not exist. The content works as a cheat sheet but fails as a progressively-disclosed skill.
Suggestions
Move the per-cloud deep dives and Terraform blocks into reference files (e.g. references/aws-cost-optimization.md, references/azure.md, references/gcp.md) and keep SKILL.md a concise overview with clearly signaled one-level-deep links.
Fix or remove the dangling references — create the referenced files or drop the pointers, and correct the inconsistent "resources/" vs "references/" directory name (line 23).
Add explicit validation/checkpoint steps to the framework and checklist (e.g. verify utilization before downsizing, confirm budget alert delivery after applying) since several listed actions are destructive or batch operations.
Trim the generic "Instructions" boilerplate and the repeated tagline so the body opens directly with cost-specific guidance.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is mostly bullet lists, concrete savings percentages, and executable Terraform, but it carries non-minor boilerplate — the repeated tagline (line 11), the generic "Instructions" section ("Clarify goals... Apply relevant best practices and validate outcomes") that applies to any skill, and tautological "Do not use" bullets — fitting the score-3 anchor of mostly efficient with unnecessary explanation that could be tightened. | 3 / 5 |
Actionability | It provides concrete, copy-paste-ready Terraform for S3 lifecycle, budget alerts, tagging locals, and auto-scaling plus specific instance types and percentages, but the Azure and GCP sections are bullet-only with no code and there are no CLI/analysis commands, leaving minor gaps versus the score-5 anchor. | 4 / 5 |
Workflow Clarity | A 4-phase Cost Optimization Framework and an end checklist give a sequence, but validation checkpoints are only implicit and the generic "validate outcomes" carries no concrete verification; the checklist includes destructive/batch actions (delete unused EBS/EIPs/snapshots, downsize) with no validation step, which per the rubric caps workflow clarity at 3. | 3 / 5 |
Progressive Disclosure | No bundle directories exist, yet the body points to non-existent files including a misnamed path ("resources/implementation-playbook.md" on line 23 vs "references/" elsewhere) and dangling "references/tagging-standards.md" and "assets/cost-analysis-template.xlsx"; meanwhile ~280 lines of per-cloud deep dives and four large code blocks that belong in separate reference files are all inlined in SKILL.md — broken references plus monolithic inlining fit the score-2 anchor. | 2 / 5 |
Total | 12 / 20 Passed |