Loading ToolBench
MCP server providing access to the Skillist skill registry, project management, and skill installation tools
Server has 7 tools with basic schemas and descriptions, but significant gaps limit production readiness. All tools have descriptions (10-60 chars, mostly short), and input schemas are present with type definitions. However, descriptions are generic and lack LLM-optimized guidance on WHEN to use each tool or HOW to chain them. Parameter descriptions are minimal or absent. Output schemas are not documented. Error handling is not visible in the provided code. Tool naming follows verb_noun convention (registry_search, registry_get_skill, add_skill_to_project) but lacks clarity on mutual exclusivity and parameter relationships. The add_skill_to_project tool has optional parameters (skillId vs externalUrl) with undocumented dependencies. No evidence of pagination support, result limits, or recovery guidance.
Add a skill or external item to a project
List projects accessible to the authenticated user
List skills and items in a specific project
Get available facets for filtering registry skills
Get detailed information about a specific skill from the registry
Get installation help and commands for a skill
Search the Skillist skill registry
Descriptions are too short (10-60 chars) and lack LLM-optimized guidance. 'Search the Skillist skill registry' does not explain WHEN to use this vs registry_get_skill, what fields are returned, or how results chain to other tools.
Parameter descriptions are missing or minimal. 'query' in registry_search has a description, but most other parameters lack guidance on format, constraints, or examples. LLMs cannot infer whether orgSlug is 'my-org' or 'my_org'.
add_skill_to_project has mutually exclusive parameters (skillId vs externalUrl) with no documented dependency or validation guidance. LLMs will not know which to provide or what happens if both/neither are supplied.
Inferred effective spec: 2026-07-28+.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | D | 50 | 2026-07-28+ | v2 |
Output schemas are not documented in the provided code. LLMs cannot plan downstream tool calls or extract required fields (e.g., does registry_search return skill_id, orgSlug, repo for chaining to registry_get_skill?).
No evidence of pagination, result limits, or error recovery guidance. registry_search and project_skills may return large result sets that blow context windows. No limit parameter or pagination cursor visible.