API Usage Dashboard
Show developers what they are calling, how often, and what is failing.
What it adds
Request volume, error, and latency reporting for an account's API traffic, broken down by key and endpoint.
What your agent is told to do
6
What your agent is told to do
6-
1
Meter every API request at one point in the stack and attribute it to an account, a workspace, and a key. Counting in several places produces numbers that never agree.
-
2
Break results down by endpoint, status class, and latency percentiles. An average latency hides the tail that is actually hurting people.
-
3
Deduplicate client retries and internal proxy hops so a single logical call is not counted three times.
-
4
Show the rate-limit window, the limit, and remaining quota with the reset time, matching the values the API returns in its headers exactly.
-
5
Do NOT display request or response bodies here. Show method, path template, status, and timing — a usage screen is not a log viewer, and payloads carry customer data.
-
6
Key creation, scopes, and last-used are owned by API Key Management; limit enforcement and rejection behaviour by Rate Limiting; product analytics by Analytics Dashboard. Link to them instead of restating them.
Edge cases it handles
6
Edge cases it handles
6- Group by path template, not by raw URL, or every record ID becomes its own endpoint row.
- Recent buckets are incomplete. Label the current window as partial rather than showing an apparent cliff.
- A deleted or rotated key still has history. Attribute past usage to it without resurrecting the key.
- 4xx caused by the caller and 5xx caused by you must be separated. Merging them makes the developer debug your outage.
- Metering must not sit on the request's critical path. Record asynchronously so instrumentation cannot slow or fail the API.
- High-volume accounts need pre-aggregated rollups; scanning raw request records at read time will be the slowest page in the product.
Definition of done
8
Definition of done
8- Usage is attributable to account, workspace, and individual key.
- Retries and internal proxy calls are counted once.
- Breakdowns include endpoint, status class, and latency percentiles.
- Displayed rate-limit windows and remaining quota match the API's own headers.
- No request or response payloads appear in the dashboard.
- The dashboard reads from aggregates and stays fast at high request volume.
- The feature matches the existing design system.
- No existing functionality is broken.
Related features
AI Cost Budgets
AI Cost Budgets
Cap what AI features are allowed to spend before the bill arrives.
What it does
Monetary spending limits on AI work, scoped by workspace, feature, and time period, enforced before a run starts.
How it works
- 1 Find every place the app calls a model and route all of them through one accounting point that records estimated and actual spend against a scope. A budget that only covers the chat feature is not a budget.
- 2 Estimate the cost of a run from the size of its input before dispatching it, and refuse anything that would exceed the remaining budget on its own.
- 3 Reserve the estimate against the budget when the run starts, then reconcile to the real usage figures when it finishes, releasing whatever was over-reserved.
Copy the prompt
No account needed
Add this feature to my app:
https://addthisfeature.com/x/ai-cost-budgets
Multi-Model Routing
Multi-Model Routing
Send each AI request to the right model using rules you can read and test.
What it does
A deterministic routing layer that picks a model per request from task type, context size, latency budget, and data sensitivity.
How it works
- 1 Express routing as explicit, ordered rules over inputs the app can measure: task type, estimated context size, latency budget, and the sensitivity classification of the data involved. A rule set that can be read line by line can be reviewed and tested.
- 2 Make routing deterministic. The same inputs must always produce the same route, so a bad output can be reproduced and a rule change can be evaluated. Randomised or load-based selection turns every incident into guesswork.
- 3 Classify data before routing and refuse to route restricted content to any destination not approved for it. This check is a hard block, not a preference, and it must run before the request is assembled.
Copy the prompt
No account needed
Add this feature to my app:
https://addthisfeature.com/x/multi-model-routing
SEO Setup
SEO Setup
Make your app findable — titles, meta, Open Graph, sitemap, robots.
What it does
The baseline SEO and social-preview setup every public app should have, and most skip.
How it works
- 1 Give every public page a unique, descriptive title and meta description. Find the app's layout and add a mechanism for each page to set them.
- 2 Add Open Graph and Twitter Card tags so shared links render a preview instead of a bare URL.
- 3 Generate a sitemap.xml covering every public, indexable page, and a robots.txt pointing at it.
Copy the prompt
No account needed
Add this feature to my app:
https://addthisfeature.com/x/seo-setup
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.