No user-submitted related servers found.
Scored across 14 tools
Most tools map to clearly distinct actions (pause, resume, delete, list, get, test, history, report). The main friction is between trigger_update, trigger_tune and trigger_draft, which all modify a trigger; the descriptions do explicitly partition them (tune = quieter only, draft+replace = code/mode changes), so an agent can resolve it, but it takes careful reading. trigger_test also partly overlaps the trial run performed by trigger_draft.
Thirteen of fourteen tools follow a predictable trigger_<verb> pattern in snake_case. The exceptions are trigger_recipes (a noun/listing verb) and service_status, which drops the trigger_ prefix entirely, but both remain readable and unambiguous.
Fourteen tools is well within the ideal range and each one covers a distinct lifecycle stage or informational need (draft, activate, pause, resume, tune, test, history, report, recipes, service status). Nothing feels padded or redundant.
The trigger lifecycle is essentially closed: create (draft), activate, pause/resume, update/tune, test, inspect (get/list/history), report a fire, and delete, plus recipes and service health. Minor gaps remain around inspecting raw probe code/execution logs and any bulk or clone operation, but agents can work around these.
