Skip to content
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension


Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
2 changes: 2 additions & 0 deletions CHANGELOG.md
Original file line number Diff line number Diff line change
Expand Up @@ -22,6 +22,8 @@
- **Desktop: Comparator v4** — Rebuilt from terminal multiplexer to structured evaluation workbench with 4-phase workflow (Setup → Execution → Results → Judge), session history rail, SQLite persistence, and shared Comparator types
- **Marketplace: Ratings and reviews** — install counts, trust tiers, and user ratings on plugin detail pages.
- **CLI: `detect` and `init` commands** — `harness-kit detect` shows which AI coding platforms are active in the current directory; `harness-kit init` scaffolds a new `harness.yaml` interactively.
- **Desktop: Permission mode selector** — Security → Permissions now has a Task Execution Mode section with three selectable modes: **Skip All** (`--dangerously-skip-permissions`, default), **Auto** (`--permission-mode auto`, requires Claude team/enterprise/API), and **Allowed Tools** (`--allowedTools <list>`, user-curated checklist). Mode and tool list persist in localStorage and apply globally or per-harness via an expandable overrides section. A one-time first-run modal explains the active mode before the first task runs and links to the settings page. A slide-in Tool Approval panel lets users manage the allowed tools list mid-session from within the terminals view.
- **Desktop: Security → Permissions redesign** — Permission mode cards with icons, flag badges, and clear selected state. Chip remove buttons now use SVG icons. Improved section hierarchy with a clean divider between execution mode and Claude settings.json controls.

### Changed

Expand Down
Loading
Loading