The Weekly Report template makes your AI produce a status report in four steps: gather the week's inputs, draft a report a stakeholder can read in under a minute, review it so every claim traces to an input, then publish. You approve before it goes out. Reject it and it returns to Draft.
4 steps · 1 approval · 1 loop back
example · the steps are this template's real steps
For managers and operators who write the same status update every week from scattered notes and task updates.
It is a stored workflow, not a prompt. Your AI client receives one step at a time from ConvOps over MCP, does the work, and advances. The engine holds the run at the approval until a person says yes.
4 steps, in this order. Below each one is the instruction your AI receives at that step, exactly as the template stores it.
Collect the inputs for this report before drafting.
Record:
Pull from notes, task updates, or the operator's recollection. If a key area is empty, ask before assuming nothing happened.
Save the gathered inputs as a task note with task_notes_add. The next step (Draft) reads this note.
Read the task notes with task_notes_list: the note from the Gather step contains the week's inputs. Work from it.
Write the report. What good looks like:
Keep it scannable: a stakeholder should grasp the state in under a minute. State numbers plainly; do not inflate.
Save the full report draft as a task note with task_notes_add. The Review step reads this note.
Read the task notes with task_notes_list: the Draft note holds the report and the Gather note holds the source inputs. Work from them.
Review the report:
Ask the operator to approve the report for publishing. This is a checkpoint: once published it informs decisions, so the AI cannot advance until you approve. If you reject, the workflow returns to Draft.
rejected → back to Draft
Read the task notes with task_notes_list: the Review note holds the approved report. Work from it.
Publish the report to its audience: the channel, doc, or email the team uses. Save the published report as a task note with task_notes_add so the record of what was sent stays on file, then close out the task.
One step waits for a person. You answer in the chat; the engine records the verdict and the audit row records who gave it.
Once published, the report informs decisions. Your AI checks every number against the gathered inputs and that blockers are stated honestly, then you approve or send it back to Draft.
what you review
reject → back to Draft
gates the engine checks without you
Read from the template's own steps: the tools they call, the files they write, and whether they touch code.
Connect your AI client first, then install Weekly Report from the catalog in the ConvOps app. See how to connect a client
An installed template is a workflow in your workspace. Ask your AI to change it in plain words; the edit lands on the one step that needs it.
new chat
more example prompts
From the Gather step: completed work, in-progress work, blockers, metrics and next week's plan, pulled from notes, task updates or your own account. If a key area is empty, the AI asks before assuming nothing happened.
That every claim and number traces back to a gathered input, blockers are not buried, the summary matches the body and nothing important was dropped.
ConvOps can schedule recurring work with a repeat rule, so a weekly report task can start every Friday. The Review approval still waits for you before anything is published.
Any MCP client connected to ConvOps. Verified clients: Claude Code, Claude (desktop, web, Cowork), OpenAI Codex, ChatGPT, Cursor, OpenCode. It needs no repository, so a plain chat client works.