- Home
- Skills
- Testing & Debugging
- Test Follow-up
Test Follow-up
After each code change, proposes exactly which tests to add (what, where, which cases) and lets you decide
Free
Works with the AI Agents you already use
Test Follow-up
Example session with this skill installed
I've finished updating the calculateDiscount function in src/utils/pricing.ts to handle the new seasonal tiers. What's next?
- Read your context and instructions
- Compiled the test follow-up
Work on the change is complete. Shall I add tests?
- Proposal: Add
pricing.spec.tsinsrc/utils/for the change insrc/utils/pricing.ts - Cases (4): standard discount tier / peak season tier / negative input handling / zero-value transactions
- Reply "yes" to proceed, "no" to skip, or "later" to leave a follow-up TODO only
Next steps
- Reply yes to generate the Vitest suite based on your existing utility patterns.
- Reply later if you want to push the logic now but keep a
TODOin the file. - Reply no if this is a spike or exploratory change.
Connects securely to your tools. The creator never sees your data.
About this skill
What it does
Agents tend to either skip tests entirely or write them unasked. Test Follow-up sits in between: right after your agent changes production code, and before it says "done", it asks you once whether to add tests, with a concrete proposal.
- Analyzes the change to find the behavior and edge cases worth verifying.
- Follows your project's conventions: reads existing tests first to match the framework (Vitest, Jest, Playwright, ...), location, and file naming. It never introduces new tooling.
- Picks the right test type: unit tests for pure functions, component tests, E2E for page flows, one regression test for a bug fix. For refactors it says plainly when the existing passing tests may be enough.
- Short proposal, then your call: what to test, where, and which cases, in 1–3 lines. Yes means it writes and runs the tests. No means it stops and does not ask again. Later means it leaves a one-line TODO.
When it stays quiet
Docs, config, or static content changes; throwaway spike code; or when you already said "skip tests".
Who it's for
Developers who want tests to stay their decision, without having to remember to ask every time.
Does not trigger for documentation or config-only changes. Requires existing test patterns to follow project conventions. Will not propose tests if user previously opted out for the session.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before any payment.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
3 uses
Times used to date
Free forever
No account required to browse
Trust & safety
Security scan
Security scan passed 1 month ago
- Free with an Agensi account