AI Conversation Summaries
Condense a long thread into the decisions made, questions open, and next steps.
What it adds
A generated summary of a conversation, structured into decisions, open questions, and actions, shown as a draft beside the thread.
What your agent is told to do
5
What your agent is told to do
5-
1
Build the summary only from messages the viewer can currently read, resolved at generation time. Deleted, redacted, and private messages must be excluded from the input, not filtered out of the output.
-
2
Structure the output into decisions that were confirmed, questions still open, and next steps with an owner where one is stated, and require the model to leave a section empty rather than pad it.
-
3
Summarise incrementally: when new messages arrive, extend the previous summary with the new material instead of regenerating from the whole thread, and carry forward earlier decisions rather than dropping them.
-
4
Present the summary as a draft with the generation time and a regenerate control. Do not post it into the thread or notify participants automatically.
-
5
AI Record Summaries covers the fields and history of a record. This feature covers message threads; if a record has both, keep them as separate summaries rather than merging them into one block.
Edge cases it handles
8
Edge cases it handles
8- A decision and a suggestion read almost the same in a chat log. Anything not explicitly agreed must be listed as an open question, and a summary that promotes speculation to a decision is a defect.
- Attribution matters when the thread contains disagreement or commitment. Name the participant behind a decision or an action, and omit attribution rather than guessing when the thread is ambiguous.
- Incremental updates lose earlier context if the previous summary is treated as disposable. Preserve prior decisions and mark them superseded when later messages overturn them.
- Messages deleted, edited, or restricted after the fact must invalidate the summary. A quote from a retracted message surviving in a summary re-publishes what someone withdrew.
- A thread of three messages does not need a summary. Set a minimum length and say the conversation is too short rather than producing filler.
- Long threads must respect a token ceiling. State when the summary covers only part of the thread and from which point it starts.
- A truncated or malformed generation must be discarded rather than shown. Keep the previous summary in place and offer a retry.
- When the model is unavailable, the thread reads normally and the summary panel shows an unavailable state.
Definition of done
9
Definition of done
9- Summaries are generated only from messages the viewer may read at the moment of generation.
- Output is separated into confirmed decisions, open questions, and next steps, with empty sections left empty.
- New messages extend the existing summary without discarding earlier decisions.
- Deleted, edited, or newly restricted messages invalidate the summary.
- Summaries appear as timestamped drafts and are never posted into the thread automatically.
- Short threads produce an explicit not-enough-content state rather than a generated summary.
- Truncated or malformed generations are discarded and the previous summary is retained.
- The feature matches the existing design system.
- No existing functionality is broken.
Related features
AI Insight Cards
AI Insight Cards
Turn a report into a few grounded cards, each tied to a number the app calculated.
What it does
A small set of cards summarising notable movements in a report, each linked to the query and figure that produced it.
How it works
- 1 Detect candidate movements in the app first, using the report's own aggregations. The model ranks and phrases the candidates it is given; it does not go looking for them and it does not produce the arithmetic.
- 2 Bind each card to the query, metric, and period that support it, and make the card link through to the filtered view so a reader can check it in one click.
- 3 Set a materiality threshold before generation, in both relative and absolute terms, so a swing on a metric with three events does not outrank a real change on a metric with thousands.
Copy the prompt
No account needed
Add this feature to my app:
https://addthisfeature.com/x/ai-insight-cards
AI Trend Analysis
AI Trend Analysis
Explain what a chart is actually showing, in sentences backed by computed numbers.
What it does
A written summary attached to a time series or metric set, describing direction, magnitude, and comparison period.
How it works
- 1 Compute every number in ordinary code before the model is involved: the change, the rate, the comparison period, the baseline, the seasonal adjustment. The model writes prose about figures it is given and must never produce a figure of its own.
- 2 State the comparison window and the time zone in the summary text, resolved from the workspace's configured zone rather than the server's. A change described as week over week is meaningless without saying which weeks.
- 3 Instruct the model to describe what the data shows and to stop there. Do not let it assert causes, attribute movement to campaigns or releases, or predict what happens next.
Copy the prompt
No account needed
Add this feature to my app:
https://addthisfeature.com/x/ai-trend-analysis
AI Key Point Extraction
AI Key Point Extraction
Reduce long content to the decisions, facts, and requests that actually matter.
What it does
A short list of the salient points in a long document or thread, each tied back to the passage it came from.
How it works
- 1 Apply this where length is the problem: long threads, meeting notes, call transcripts, lengthy tickets, multi-page documents. Content that already fits on a screen does not need extracting.
- 2 Classify each point by kind — decision, fact, risk, request, open question — so a reader can scan for the category they came for instead of reading a flat list.
- 3 Carry a reference to the source passage with every point and let the reader jump to it. A point nobody can verify is a claim, not a summary.
Copy the prompt
No account needed
Add this feature to my app:
https://addthisfeature.com/x/ai-key-point-extraction
How it works
-
1
Copy the link
Grab the Markdown instruction URL for this feature.
-
2
Give it to your AI
Paste it into Claude Code, Cursor, v0, Lovable — whatever you build with.
-
3
It inspects, then implements
Your agent reads your existing app first, then adds the feature to fit it.
Works with your stack
These instructions are written to adapt. They tell the agent to detect your framework, match your existing design system, and reuse what you already have — rather than assuming a particular stack.
Need it tighter than that? Customize the feature and tell it exactly what you're running.