Unused Token Detection
Find the design tokens nothing references any more, before they harden into permanent debt.
What it adds
A report of every theme and design token in the codebase alongside where, or whether, it is still referenced.
What your agent is told to do
5
What your agent is told to do
5-
1
Find the files that define the app's tokens — colours, spacing, radii, type steps, shadows, motion durations — and treat that set as the authoritative list to check against.
-
2
Resolve every reference the codebase makes, in stylesheets, component source, inline styles, and any theme configuration, and record the count and the locations against each token.
-
3
Report three distinct outcomes rather than one: referenced, referenced only by other tokens, and referenced by nothing at all. The middle case is the one people delete by mistake.
-
4
Make the report a repeatable check that can be run on demand and read in a diff, not a one-off list pasted into an issue that is stale within a week.
-
5
Do not delete anything as part of this feature. Detection and removal are separate acts, and an automatic sweep that removes a token a customer theme still sets will break that theme silently.
Edge cases it handles
6
Edge cases it handles
6- Tokens assembled at runtime — a name built from a variant string, a colour picked by lookup from a data value — will look unused to any scan that only matches literal text. Detect the patterns that build names dynamically and mark everything they could reach as uncertain rather than unused.
- A token that is deliberately deprecated but still shipped for compatibility is not the same as a token nobody ever used. Carry the deprecation status through to the report so the two are never merged into a single number.
- Aliases must be traced to the end of the chain before any judgement is made. A base token referenced only by an alias that is itself referenced by a live component is in active use, and reporting it as dead is the most common way this check causes an outage.
- The scan must cover every theme, every brand, and every package in the repository. A token that is unused in the default theme and load-bearing in the dark or high-contrast one is used.
- Tokens consumed by consumers outside this repository — an embed, a published package, a documented customer theming surface — cannot be judged from this codebase alone and must be flagged as externally visible.
- Tokens referenced only by tests, stories, or documentation fixtures should be reported separately, since that is evidence of abandonment rather than of use.
Definition of done
8
Definition of done
8- Every token defined in the app appears in the report with its reference count and locations.
- Alias chains are resolved to their end before a token is classified.
- Deprecated, unused, and dynamically referenced tokens are three separate categories in the output.
- The scan covers all themes and all packages, not only the default theme.
- The check can be re-run on demand and produces a stable, diffable result.
- No token is removed by the feature itself.
- 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
Model Selection
Model Selection
Let each AI task run on the model that suits its quality, speed, and cost needs.
What it does
A per-task model choice, drawn from the models the app already has configured, with capability filtering and safe defaults.
How it works
- 1 Enumerate the models the app already has access to and record what each one can actually do: context capacity, whether it can return the structured output the task requires, whether it supports the tools the task calls, and its relative cost and speed.
- 2 Offer only the models that satisfy the task's requirements. A task that needs structured output must not list a model that cannot reliably produce it, because the failure appears later as malformed responses rather than as an unavailable option.
- 3 Store the choice against the specific task, not as one global setting. A single default forces a summarisation task and a classification task onto the same tier when they have opposite needs.
Copy the prompt
No account needed
Add this feature to my app:
https://addthisfeature.com/x/model-selection
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.