Loading ToolBench
MCP server for LiquidSoap 2.4.0 documentation and script assistance
This server provides 7 tools for LiquidSoap documentation and script assistance. All tools are READ_ONLY and present basic schemas with descriptions. However, multiple critical gaps limit quality: (1) Most tool descriptions are surface-level and lack actionable context for when/why to use them. (2) Input schemas are present but sparse, many lack detailed parameter descriptions explaining format, range, or constraints. (3) No output schemas are documented, forcing LLMs to guess the structure of returned data. (4) Parameter descriptions are often generic (e.g., 'Search term' for search_functions) without examples of valid inputs or expected behavior. (5) No error handling guidance is provided, tools will fail silently or return generic errors. (6) The validate_script_syntax tool lacks specificity about what 'basic syntax validation' means or what best practices it checks. The tool set is functionally coherent (all read-only documentation helpers), but the definitions lack the precision and explicitness required for reliable LLM usage.
Get the changelog and breaking changes for LiquidSoap 2.4.0
Fetch a specific section of the LiquidSoap 2.4.0 documentation
Get code examples for common LiquidSoap tasks
Get the LiquidSoap version this server supports (2.4.0)
List all available documentation sections with descriptions
Search for functions or operators in the LiquidSoap API reference
Provide basic syntax validation and best practices for a LiquidSoap script
Output schemas are completely undocumented. Tools return text strings or structured data, but LLMs have no declared schema to understand the response format. This forces LLMs to infer structure and wastes tokens on parsing.
Parameter descriptions lack actionable detail. 'Search term (function name, keyword, or description)' for search_functions is vague, what happens if the term doesn't match anything? What format should keywords be in? How many results are returned?
validate_script_syntax description is ambiguous. 'Provide basic syntax validation and best practices' doesn't specify: What validation rules apply? What constitutes a best practice error vs. a syntax error? What does the output look like (pass/fail boolean, list of errors, scored feedback)?
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-22 | D | 51 | 2026-07-28+ | v2 |
| 2026-03-09 | F | 0 | - | v1 |
No error handling guidance. If search_functions returns 'No functions found', there's no hint about whether the user should try a different query, browse list_sections, or check the LiquidSoap version. Tools silently fail without recovery paths.
get_documentation max_length parameter lacks clear semantics. 'Maximum length of returned content in characters (default: 50000)', but what happens if the section exceeds this? Is it truncated? Summarized? What if the user needs more?
Tool descriptions do not explain when to use each tool vs. alternatives. Is search_functions for finding function signatures, or for discovering function capabilities? Should users call list_sections first to explore? The decision logic is left to the LLM.