No user-submitted related servers found.
Scored across 15 tools
Each tool maps to a distinct resource+action (playbook get/list/create, skill list/get/reference, memory read/write/search, canvas list/read/write), so an agent can generally pick correctly. The main friction is that several descriptions spend more words pointing at sibling tools that are absent from the set (get_memory_history, patch_canvas_section, call_connected_tool), which muddies boundaries without creating true overlap.
All 15 tools use a consistent snake_case verb_noun pattern (get_playbook, list_skills, write_memory, read_canvas, search_memory). Verb choice is predictable per action family (list/get/read for retrieval, write/create/update for mutation), with no camelCase or stylistic mixing.
15 tools is within a reasonable band and each entry is a plausible slice of the playbook domain. The count is slightly high for what is effectively a read-heavy partial surface, and the presence of a meta tool (find_tools) alongside the concrete operations adds a bit of redundancy.
The surface is a fragment of the described domain: there is create_playbook but no update_playbook or delete_playbook, skills are read-only with no create/attach, memory has no delete/archive/history, and canvas lacks section patching, TOC, and locking. Worse, descriptions explicitly instruct agents to use tools that are not present (update_playbook, get_canvas_toc, patch_canvas_section, get_memory_history, call_connected_tool), which will produce dead ends.
