The Meeting Follow-Up template makes your AI turn a call into a sent follow-up: capture who attended and what was said, extract decisions, action items and the next step, draft the message, then check it against what was agreed. Sending is irreversible, so it stops for your approval. Reject it and it redrafts.
5 steps · 1 approval · 1 loop back
example · the steps are this template's real steps
For sales reps, account managers and founders who take client calls and owe a recap with next steps after each one.
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.
5 steps, in this order. Below each one is the instruction your AI receives at that step, exactly as the template stores it.
Capture what happened in the meeting before structuring anything.
Record:
If the record is thin, ask the operator for the missing pieces: a vague capture produces a vague follow-up.
Save the raw meeting record as a task note with task_notes_add. The next step (Extract) reads this note.
Read the task notes with task_notes_list: the note from the Capture step contains the raw record. Work from it.
Extract the structure. What good looks like:
Pull only what was actually said: do not invent commitments.
Save the extracted decisions, actions, and next steps as a task note with task_notes_add. The Draft Follow-Up step reads this note.
Read the task notes with task_notes_list: the Extract note holds the decisions, actions, and next steps, and the Capture note holds the attendees and context. Work from them.
Write the follow-up message:
Keep it scannable: the reader should see the next step in seconds.
Save the full follow-up message as a task note with task_notes_add. The Send Review step reads this note.
Read the task notes with task_notes_list: the Draft Follow-Up note holds the message and the Extract note holds what was agreed. Work from them.
Check the follow-up before sending:
Ask the operator to approve the message for sending. This is a checkpoint: sending is irreversible, so the AI cannot advance until you approve. If you reject, the workflow returns to Draft Follow-Up.
rejected → back to Draft Follow-Up
The follow-up is approved and sent. Confirm the final message and the extracted action items are saved as task notes, so the record of what was agreed and 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.
Sending is irreversible. The AI checks the recap, every owner and date, and the names against what was agreed, then you approve or send it back to Draft Follow-Up.
what you review
reject → back to Draft Follow-Up
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 Meeting Follow-Up 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
The Extract instruction pulls only what was actually said and does not invent commitments. Send Review checks the recap against the extracted decisions before you approve.
Your AI client sends or hands you the final message, using whatever tools it is connected to. ConvOps holds the run at Send Review until you approve.
The final message and the extracted action items are saved as task notes, so what was agreed and what was sent stay on file for the next person.
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.