Loading ToolBench
Dependency license + SBOM gate with offline CVE enrichment, developer-CLI first
Single tool 'licenselens_scan' has a basic description (67 chars) and minimal schema. Input parameter 'target' lacks type specification in visible code (only string type inferred from FastMCP decorator). No output schema documented. No error handling guidance, no parameter constraints, no recovery paths. Tool does one thing (scan) but lacks the rigor expected for production agent integration. Description is adequate but terse; no guidance on when to use, what to expect, or how to handle failures.
Dependency license + SBOM gate, developer-CLI first. Returns JSON findings.
No output schema documented. LLM cannot predict response structure, fields, or types. Breaks downstream tool chaining and forces unstructured parsing.
Parameter 'target' lacks description. LLM cannot determine if it accepts file paths, directory paths, or both; no guidance on format or constraints.
No error handling guidance. Tool description does not explain failure modes (file not found, invalid format, unresolvable licenses) or recovery steps.
No parameter constraints or validation hints. LLM cannot infer valid input formats, acceptable file types, or size limits.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | F | 49 | 2026-07-28+ | v2 |
Tool description does not state what 'JSON findings' contains. LLM must guess at response structure (license violations? SBOM format? severity levels?).