Content
87%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
A tightly written, highly actionable catalog of business-logic attack primitives with a useful decision tree for ordering them. Its only gap is the absence of explicit validation/stop gates on the batch probe loops, which caps workflow clarity.
Suggestions
Add an explicit success check to the probe loops (e.g., stop and report when a response contains a flag/admin indicator) so each technique has a clear validation gate rather than relying on implicit grep inspection.
Convert the trailing 'Anti-pattern (real-world evidence)' anecdote into a one-line directive; the specific elapsed-time figures are motivational padding that does not change what Claude should do.
In the Decision Tree, note what to do when a probe returns an unexpected status code (e.g., 500 or redirect) so error recovery is explicit rather than implied.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is overwhelmingly executable commands with only brief, purposeful tactical context per section; it assumes Claude's competence and avoids explaining what 2FA or curl is, with every section earning its tokens. | 3 / 3 |
Actionability | Provides copy-paste-ready bash with real wordlist paths (/usr/share/wordlists/rockyou.txt, seclists), concrete curl loops, and specific flags — fully executable guidance rather than pseudocode or abstract direction. | 3 / 3 |
Workflow Clarity | A Decision Tree sequences the six approaches, but the batch curl loops and hydra brute-force lack explicit validate/stop checkpoints (e.g., 'halt when a flag appears'), so per the batch-operations guideline workflow clarity is capped at 2. | 2 / 3 |
Progressive Disclosure | No bundle files exist and none are needed; the single SKILL.md is cleanly organized into six numbered sections plus a decision tree, giving easy navigation without a monolithic wall of text. | 3 / 3 |
Total | 11 / 12 Passed |