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 highly actionable, well-sequenced offensive-testing skill: every step is executable, the admin-baseline-then-replay methodology is explicitly validated with a checklist and verification script, and there is no wasted prose. The only structural improvement would be offloading the framework table and transport-specific sections into reference files to slim the main SKILL.md.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is dense and lean — almost entirely executable commands, terse lists ("Look for: method returning 200/204 where others return 403"), and tables, with no padding or explanation of concepts Claude already knows. Minor trimmable framing (intro paragraph, Detection Notes, ATT&CK mapping) keeps it just below anchor 5. | 4 / 5 |
Actionability | Fully executable, copy-paste-ready curl, Python, and grpcurl snippets with placeholders cover the common cases: admin baseline and replay diff, method alternation loop, encoding/trailing-slash alternation, GraphQL mutations with alias batching, background job approval, and header trust bypass. The framework weaknesses table gives concrete, named patterns per framework. | 5 / 5 |
Workflow Clarity | A clear numbered sequence (recon and actor x action matrix, admin baseline first to rule out a universally broken action, lower-privilege replay with a stated win condition, then method/transport/GraphQL/job/header variations) is capped by an explicit Verification section with a 3-part confirmation checklist and an executable verification script. Validation checkpoints for the destructive/batch operations are present, so the workflow-clarity cap does not apply. | 5 / 5 |
Progressive Disclosure | No bundle files exist, so the skill is a single self-contained file; it is well-sectioned with clear headers and easy navigation. At ~240 lines, content such as the framework weaknesses table, gRPC testing, and detection notes could be split into reference files, which keeps it below anchor 5's ideal split; structure is far above anchor 3's disorganized inline bulk. | 4 / 5 |
Total | 18 / 20 Passed |