No user-submitted related servers found.
Scored across 65 tools
Despite very detailed descriptions, several families blur together: render_session/export_session/decision_record all output a session, save_decision_map/save_decision_card both persist decision artifacts, and the calibration cluster (calibrate, prediction_calibration, calibrated_consult_journal, advisor_weights, mirror_report) is hard to tell apart. loop_status/pending_outcomes/board_status all surface pending outcomes, adding further overlap. An agent must read carefully to avoid misselection.
Names are consistently snake_case and grouped by prefix families (federation_*, catalog_*, ollama_*, calibrated_consult_*), which aids readability. However the verb/noun convention is mixed: some are noun_status (board_status, job_status), some verb_noun (build_advisor, seed_council), and many are bare nouns/verbs (cite, retrieve, calibrate, premortem, doctor, stability). No single predictable pattern across the set.
65 tools is far beyond a manageable surface and well past the 25+ threshold for 'too many'. The domain (multi-brain federation, PD catalog, decision science, calibration, sessions) is genuinely broad, which explains some sprawl, but many tools could be consolidated into submodes. The count alone makes discovery and correct selection difficult.
Coverage is extensive: onboarding/scaffolding, corpus building, catalog fetch/verify, federation orchestration, decision mapping/calculation, prediction and overreliance calibration, and session rendering/export are all present. Minor gaps exist (e.g. no delete/invalidate operations for advisors or saved decision artifacts), but core lifecycle paths appear well covered.
