No user-submitted related servers found.
Scored across 65 tools
Multiple tools occupy nearly the same conceptual niche: wiki_get_flow_context, wiki_get_briefing, and planning_context all return pre-planning context for a flow, while search and wiki_search both search flows, and devflow_init overlaps with agent_session_create. The descriptions are detailed, but an agent selecting among 65 tools will frequently face unclear boundaries.
Most subdomains follow a clear domain-prefix pattern (flow_, task_, doc_page_, release_, wiki_, adr_, knowledge_, agent_session_), which keeps the set readable. However, the pattern breaks often: flow_comments_get is object-first, wiki_backlinks and wiki_graph_neighbors are nouns rather than get_* verbs, devflow_token_emit vs devflow_tokens_list mix singular/plural, and standalone names like search, planning_context, pending_work, and ideas_get escape the convention.
65 tools is an extreme surface for a single MCP server, far beyond the 50+ threshold. Even if each subdomain is individually justified, the aggregate tool count forces agents to parse a huge namespace and makes misselection much more likely.
The server covers the full flow lifecycle, tasks, sessions, doc pages, releases, wiki/ADR reads and writes, knowledge-draft workflows, and discipline tokens, so core workflows have no dead ends. Minor gaps exist: no release delete, no single-task get, no way for agents to create flow comments, and no direct wiki page create for patterns/runbooks except via knowledge drafts.