AI Agent Actions with Approval
Let the assistant draft app actions and run them only after the user approves the exact plan.
What it adds
A plan-and-approve step between an assistant's proposed actions and their execution, showing every target and change before anything runs.
What your agent is told to do
5
What your agent is told to do
5-
1
Split the flow in two. Producing a plan must have no side effects at all, and execution must run only from an approved, stored plan rather than from the conversation that produced it.
-
2
Render the plan in the app's own language: which records are affected, which fields change from what to what, and every external side effect such as an email, a charge, or a webhook. Nothing may execute that was not on the screen the user approved.
-
3
Re-check permissions and the current state of every target at execution time, not at planning time. A plan drafted two minutes ago may reference a record that has since been edited, deleted, or locked.
-
4
Invalidate the approval whenever the plan changes. Any edit, re-plan, or change in the underlying records requires a fresh approval covering the new plan, and an approved plan should expire after a short window.
-
5
The catalogue of actions the model may invoke, their argument schemas, and the call limits belong to AI Tool Calling Framework. Use that catalogue for the actions in a plan rather than defining a second, parallel set of capabilities here.
Edge cases it handles
8
Edge cases it handles
8- Every target, every field-level before-and-after, and every outbound side effect must be visible before approval. A plan that summarises itself as updating some records has not been approved in any meaningful sense.
- Permissions and record state are revalidated at execution. If a target moved out of scope between approval and execution, skip that step, mark it clearly, and do not silently substitute another record.
- A plan that changes for any reason loses its approval. Re-approving a modified plan must show what changed since the last approval, not present it as if it were new.
- Steps must be individually idempotent and keyed, so a retry after a partial failure does not send a second email or apply a change twice. On partial failure, report exactly which steps completed, which did not, and what state the data is now in.
- Arguments the model generated but the interface did not display must never be executable. If a value cannot be rendered for inspection, the plan is not approvable.
- Long-running plans need a visible progress state and a way to stop between steps. Stopping mid-plan must leave a clear record of the boundary reached.
- Approval must be attributable to a specific person and recorded in the audit trail alongside the plan as approved, so the plan text cannot be reconstructed differently later.
- If the model is unavailable or returns malformed structured output, the user must still be able to perform the same actions through the ordinary interface. Nothing should be reachable only through the assistant.
Definition of done
9
Definition of done
9- Planning produces no side effects, and execution runs only from a stored approved plan.
- The approval screen shows every target, every field change, and every external side effect.
- Permissions and target state are revalidated at execution time and out-of-scope steps are skipped and reported.
- Any change to a plan invalidates the approval and requires re-approval showing the diff.
- Steps are idempotent and partial failures report exactly what did and did not run.
- No argument can be executed that was not displayed to the approver.
- Approvals are attributed to a person and recorded with the plan as approved.
- The feature matches the existing design system.
- No existing functionality is broken.
Related features
Streaming AI Responses
Streaming AI Responses
Show AI output as it is generated so long answers do not look frozen.
What it does
Incremental rendering of model output as it arrives, with a persisted authoritative copy once the stream closes.
How it works
- 1 Find the AI surfaces where the user waits on a long response and stream those. Short classification or extraction calls do not benefit and should stay as plain requests.
- 2 Buffer incoming bytes until they form complete characters and complete structural units before rendering, so partial output never appears as broken glyphs or half-open markup.
- 3 Treat the streamed text as provisional. When the stream closes, persist the provider's final message as the authoritative record and reconcile what is on screen against it.
Copy the prompt
No account needed
Add this feature to my app:
https://addthisfeature.com/x/streaming-ai-responses
AI Support Triage
AI Support Triage
Summarise each incoming support request and estimate how urgent it really is.
What it does
An automatic summary, urgency estimate, and likely problem area attached to every inbound support request, with the evidence behind each.
How it works
- 1 Run triage as a background job on arrival, using the app's existing queue and retry behaviour. The ticket must be visible and workable in its raw form the moment it lands, whether or not triage has finished.
- 2 Produce three separate outputs — a short summary, an urgency level, and a likely problem area — and store each with the quoted source statements from the ticket that support it, so a human can check the reasoning.
- 3 Define urgency in terms of business impact: who is blocked, how many, whether money or data is at risk. Compute it from what the ticket describes, not from how the writer sounds.
Copy the prompt
No account needed
Add this feature to my app:
https://addthisfeature.com/x/ai-support-triage
AI Suggested Support Answers
AI Suggested Support Answers
Give support agents a grounded first draft instead of a blank reply box.
What it does
A draft reply composed for the agent from the ticket, the customer's account state, and the approved internal knowledge, with sources attached and no ability to send itself.
How it works
- 1 Ground every draft in retrieved material: the ticket thread, the account's real state, and articles from the approved knowledge set. Attach the sources used to the draft so the agent can open and check each one.
- 2 Load the draft into the agent's normal reply editor, unsent and fully editable. There is no path in this feature that sends a message to a customer without an agent pressing send.
- 3 Define the commitments the app is not allowed to make in a draft — refunds, credits, delivery dates, guarantees of a fix — and strip or refuse any draft containing them, leaving that part for the agent to write.
Copy the prompt
No account needed
Add this feature to my app:
https://addthisfeature.com/x/ai-suggested-support-answers
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.