Observed arrival · 2026-10-30
Portia: tests before agent-built features
Portia presents a Claude Code skill and CLI for turning an epic into human-approved failing tests, parallel coding-agent work, and a single pull request.
○Open
⊠Login
$Paid
†Ads
✦Pretty
●Pro
◎Niche
◉Human
⚑Risk
ƒJS
- For
- Engineering teams coordinating Claude Code workers
- Worth noticing
- A replayed example run shows parallel waves, a failing gate retry, and an approval for changes outside a child task’s paths.
Field notes
The page describes a “spine” that writes acceptance specs before child tasks exist; those specs are shown failing by design until the corresponding work lands. It says each child records the scenario and test it proved, while an epic-level ledger is rebuilt from the tracker, GitHub, and run log after each landing. The example also depicts a gate, a retry, and a human-approved change outside a child’s paths.
Observed signals
Read the marks
Editorial observations of this landing page, not a rating.
○OpenPublic substance visible
$PaidCommerce or pricing visible
✦PrettyNotable craft visible
●ProPolished or operationally mature
◎NicheUnusually specific use
ƒJavaScriptBrowser-side code central
One card from the complete issue