research: QA agent design — what does automated agent testing look like in practice? #52

Closed
opened 2026-06-28 19:26:53 +00:00 by Claude · 0 comments
Collaborator

Context

Open design question. Before the QA role agent can be built, the testing model needs to be defined.

Questions to resolve

  • What constitutes an automated agent test? (eval vs integration vs behavioral?)
  • Where do agent tests live relative to skill evals?
  • What tooling/framework runs them?
  • How does this connect to plugins/kyberforge/tests/evals/?

Output

Design note (docs/notes/qa-agent-design.md) and, if a hard architectural decision is reached, an ADR.

## Context Open design question. Before the QA role agent can be built, the testing model needs to be defined. ## Questions to resolve - What constitutes an automated agent test? (eval vs integration vs behavioral?) - Where do agent tests live relative to skill evals? - What tooling/framework runs them? - How does this connect to `plugins/kyberforge/tests/evals/`? ## Output Design note (`docs/notes/qa-agent-design.md`) and, if a hard architectural decision is reached, an ADR.
Claude added this to the Skills & Agents milestone 2026-06-28 19:26:53 +00:00
Claude added the Kind/Feature
Priority
Medium
3
labels 2026-06-28 19:26:53 +00:00
Sign in to join this conversation.