Check your cron jobs, recent runs, workflows and webhooks, and run a job now when you confirm.
Scored across 11 tools
Most tools have clearly distinct purposes and the descriptions explicitly add 'Not for...' boundaries (e.g. get_job vs get_job_runs vs run_job, get_workflow_run vs list_workflow_runs). The main soft spot is the pair open_croncool and show_project_overview, which both surface project/job info and could be confused with list_projects/list_jobs, though the interactive-vs-read-only framing helps.
Strong, predictable verb_noun snake_case pattern throughout (get_job, list_jobs, run_job, list_workflow_runs, show_project_overview). The slight deviation is open_croncool, which is a bare verb plus product name rather than verb_noun, but it reads as expected branding.
11 tools is well-scoped for a cron/workflow domain spanning jobs, projects, workflows, webhooks, and runs. Each tool maps to a distinct read or action, with no bloat.
The surface is heavily read-oriented plus a single run_job trigger; there is no create_job, update_job, delete_job, or enable/disable toggle despite enabled state being a core attribute. Webhook subscriptions are list-only with no re-enable or change, and workflows cannot be created or triggered, leaving notable lifecycle gaps.
Use this when the user wants to read the scheduling configuration and latest execution metadata of one CronCool job selected by its exact identifier. Returns the job's name, cron expression, type, target origin, HTTP method, content type, enabled state and last run status and time, or not-found. Not for running the job.
Inputs: jobId (required). Returns: job, status
Use this when the user wants to read execution history for one CronCool job to inspect success, failure, HTTP status and duration. Returns recent runs newest first with success or failure, HTTP status, duration and time (no response bodies), 20 by default and 100 at most. Not for starting a run.
Inputs: jobId (required), limit. Returns: jobId, executions
Use this when the user wants to read one CronCool workflow execution and a bounded page of its step statuses for troubleshooting. Returns the run's status, timings and error code plus its steps with status, attempt, timings and error code, 50 steps by default and 200 at most. Not for listing runs.
Inputs: runId (required), projectId (required), stepLimit. Returns: run, steps, status, stepsTruncated
Use this when the user wants to list scheduled jobs in CronCool to inspect schedules, enabled state and the latest execution status. Returns each job's id, name, schedule, target origin, method, enabled state and last run status and time, 20 per page by default, with a cursor for more. Not for a job's full run history.
Inputs: limit, cursor, projectId. Returns: jobs, hasMore, nextCursor
Use this when the user wants to list projects accessible to the connected CronCool account, including identifiers needed to select a project. Returns each project's id, name and creation time, 50 by default and 100 at most, with a flag when more exist. Not for listing jobs.
Inputs: limit. Returns: hasMore, projects
Use this when the user wants to read CronCool webhook subscription configuration and recorded delivery failures. Returns each subscription's endpoint origin, events, active state, consecutive failures, auto-disabled flag and last delivery time, 20 by default. Works only with a signed-in user session, not an API key. Not for changing or re-enabling a webhook.
Inputs: limit, projectId. Returns: subscriptions
Use this when the user wants to list executions of a durable workflow in CronCool, with optional workflow-name and status filters. Returns each run's id, workflow name, status, timings and error code, newest first by default, 20 per page with a cursor, filterable by workflow and status. Not for one run's steps.
Inputs: projectId (required), sort, limit, cursor, +2 more. Returns: runs, hasMore, nextCursor
Use this when the user wants to list durable workflow definitions and run-count summaries for one CronCool project. Returns each workflow's name, run counts by status and last run time for one project, 50 by default and 100 at most. Not for individual runs.
Inputs: projectId (required), limit. Returns: hasMore, workflows
Use this when the user wants to open the CronCool project-selection sidebar when the user requests an interactive project and job browser. Returns up to 100 projects to start from (no request payloads, workflow bodies, secrets or full target URLs). Not for running jobs or changing their configuration.
Inputs: no named parameters. Returns: items, truncated
Use this when the user wants to immediately execute the configured request for one CronCool job. Returns only that Croncool invoked it, not the target's response. Each call immediately sends the job's configured request again, which can write, message or bill downstream, and a repeated call sends it once more. Not for checking a run's result.
Inputs: jobId (required). Returns: jobId, status
Use this when the user wants to render a read-only overview of one CronCool project and a bounded preview of its scheduled jobs. Returns an overview card with the project's name and creation time, whether workflow hosting is set up, and up to 12 jobs with schedule, target origin and last run status. Not for a job's run history.
Inputs: projectId (required). Returns: jobs, project, jobCount, jobCountIsLowerBound
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
Debug background jobs — runs, traces, spans, schedules, queues and deployments.
Know when cron jobs and AI agents stop running: monitors, check-ins, incidents and runbooks.
Manage CronPulser from an AI agent: HTTP, Runner and TLS jobs, alerts, Runners and run history.
Manage cron jobs, machines and run history in your norc account from any AI chat.
Add one secure layer between your agents and this server.
