Skip to content

paradigmxyz/centaur / .agents/skills/scheduled-tasks

scheduled-tasks

Verified

Create, list, read, update, delete, or run Centaur scheduled tasks with the `centaur_console` Code Mode connector. Use when a user asks to schedule recurring or one-off agent work, such as "every Monday at 9 summarize #channel", "remind me daily", "run this report weekly", or to pause, change, delete, or trigger an existing task.

Scheduled Tasks

Manage recurring or ad-hoc agent work with the centaur_console connector in Code Mode (centaur-codemode exec --result-only). centaur-codemode load centaur_console prints every function and its arguments.

const cc = tools.centaur_console;
await cc.scheduled_tasks();                                   // tasks owned by this user or channel
await cc.scheduled_task({ task_id: "tsk_..." });
await cc.create_scheduled_task({ name, prompt, cron_expression, delivery_channel, enabled });
await cc.update_scheduled_task({ task_id, name, prompt, cron_expression, delivery_channel, enabled });
await cc.delete_scheduled_task({ task_id });
await cc.run_scheduled_task({ task_id });                     // queue an immediate run

Only name and prompt are required to create a task; omitted fields on update are kept.

Ownership and delivery

  • In MCP or direct-message (DM) sessions, tasks belong to the user and deliver to dm (the default when delivery_channel is omitted) or an allowed Slack channel ID.
  • In a Slack channel thread, tasks belong to that channel: they run with the channel's permissions, always post to it (omit delivery_channel), and anyone in the channel can manage them.

Schedules and prompts

  • Schedules are optional five-field cron expressions in Pacific Time. Omit cron_expression for an ad-hoc task triggered with run_scheduled_task; pass cron_expression: "" to update_scheduled_task to clear a schedule.
  • Encode recurrence only in the cron expression. Strip cadence phrases like "Each Monday" from the prompt, but keep time windows that affect the work, such as "messages from the past week".
  • Write the prompt so it stands alone: the task runs in a fresh session without this conversation's context.

After a change

Report the task ID, schedule, destination, enabled state, and next run. Do not repeat a successful mutation; if a call's outcome is unclear, check with scheduled_tasks or scheduled_task before retrying.

reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis.Stars: 471kUpdated Claude CodeCodexVerifiedInstall
diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness.Stars: 322kUpdated Claude CodeCodexVerifiedInstall
dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependenciesStars: 322kUpdated Claude CodeCodexVerifiedInstall
executing-plansobra/superpowersUse when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is availableStars: 322kUpdated Claude CodeCodexVerifiedInstall