A headless, config-as-code DAST runner that crawls an authenticated web/mobile-API surface and fires a curated active-scan ruleset, emitting deduplicated SARIF.
Single tool 'dastlite_scan' has a reasonable description (127 chars, within 10-1024 range) but critical schema gaps. Input schema visible but lacks type annotations on the 'target' parameter, only a string description provided, no 'type' field in JSON Schema. No output schema documented. Parameter description is minimal (11 chars: 'The target URL to scan'). Tool name lacks action verb prefix (should be 'scan_dastlite' or 'run_dast_scan'). No error handling guidance, no pagination support, no structured output documentation. Naming follows verb_noun loosely but 'dastlite_scan' is tool-centric rather than action-centric.
A headless, config-as-code DAST runner that crawls an authenticated web/mobile-API surface and fires a curated active-scan ruleset, emitting deduplicated SARIF.. Returns JSON findings.
Input schema lacks proper JSON Schema type definitions. Parameter 'target' has description but no explicit 'type' field, violating JSON Schema structure. Cannot validate input programmatically.
No output schema documented. Tool returns 'JSON findings' but LLM cannot predict field structure, types, or pagination. Violates pattern:tool requirement for documented return types.
Tool name 'dastlite_scan' is tool-centric, not action-centric. Should start with clear action verb like 'scan_', 'run_', or 'execute_'. Current name obscures intent for LLM tool selection.
Parameter description 'The target URL to scan' is only 22 chars and lacks format guidance. Should specify: expected URL format, authentication requirements, timeout behavior, and example constraints.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | F | 43 | 2026-07-28+ | v2 |
No error handling guidance. Tool description does not explain failure modes (invalid URL, timeout, auth failure, scan errors) or recovery steps. LLM cannot self-correct on failure.