Content
86%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 a tight, executable recon procedure: a runnable endpoint-discovery loop, a dense mechanism-to-signal identification table, and a concrete handoff output spec, organized into clean sections with no padding or over-explanation. Its only notable gaps are the absence of runnable commands for a couple of signal checks and no explicit retry/validation loop, which keep actionability and workflow clarity at 4.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is lean with no explanation of concepts Claude already knows — the intro only states purpose ("Exploit downstream needs to know exactly how to hold a session") and the signal table is dense and non-redundant, matching the 5 anchor where every token earns its place. | 5 / 5 |
Actionability | The curl status-code loop over common auth paths is real, executable bash, and the mechanism table gives concrete detection signals (`Set-Cookie`, `Authorization: Bearer eyJ...`, `SAMLRequest`), matching the 4 anchor; it falls short of 5 because probing Bearer/API-key acceptance is described as signals rather than given as runnable commands, and section 3 is guidance rather than code. | 4 / 5 |
Workflow Clarity | A clear three-step sequence (discover endpoints → identify mechanisms → record handoff output) with the inline `[ "$code" != "404" ]` filter and the "success/fail signals" spec acting as checkpoints, fitting the 4 anchor; it lacks the explicit validate-fix-retry feedback loop of the 5 anchor, but this is read-only recon so no destructive/batch cap applies. | 4 / 5 |
Progressive Disclosure | No bundle files exist (no references/, scripts/, or assets/), and the body is under 50 lines organized into three well-labeled sections with no content that needs to be split out — per the rubric's simple-skill guideline this scores 5. | 5 / 5 |
Total | 18 / 20 Passed |