TASK GUIDE · SOURCE-AWARE

Best Agent Skills for test-driven development

A focused Skill plan for defining behavior, writing the smallest test, and verifying the change.

Build this plan for my Agent ↗
Desired outcome

Make the desired behavior executable before implementation and keep the change small.

Safe working boundary

One behavior at a time; do not rewrite unrelated code to make a test pass.

01 · SKILL STACK

A minimal three-role stack

Each role has one job. The personalized builder can adjust the selection for your Agent without changing the decision logic.

01 · Core execution

tdd

Drives the red, green, refactor loop.

Main limitation
The workflow needs a functioning test runner and clear public behavior.
First risk
False confidence
Source evidence
Pinned · 885e2ca4
Inspect the Skill ↗
02 · Guardrail

create-implementation-plan

Keeps the test tied to the actual change boundary.

Main limitation
Demanding zero ambiguity can create oversized plans or false precision for exploratory work.
First risk
False certainty from generated detail
Source evidence
Pinned · 318066d2
Inspect the Skill ↗
03 · Evidence

minimal-run-and-audit

Runs the smallest relevant suite and records the result.

Main limitation
It does not choose the reproduction target, train a model, conduct broad paper analysis, or hide scientific-meaning changes.
First risk
Scientific changes hidden as setup fixes
Source evidence
Pinned · 3ab50525
Inspect the Skill ↗

02 · ACCEPTANCE

How to know the task is finished

  1. Show the new test failing for the intended reason.
  2. Make the smallest implementation change that passes it.
  3. Run the adjacent regression tests before refactoring.

03 · STOP CONDITIONS

When to stop

  • The test only mirrors implementation details.
  • Unrelated behavior changes to make the test pass.
  • The full suite is skipped after shared code changes.

04 · METHOD

How this guide was produced

SkillSignal maps a defined task to three non-overlapping responsibilities, then selects from reviewed catalog metadata. Agent fit, source state, permissions, and published scores are evaluated with deterministic rules. No LLM writes or reorders the recommendation at request time.