This release adds 4 notable features for engineering teams evaluating rollout.
✓ No known CVEs patched in this version
Topics
+13 more
Summary
AI summaryAdded a terminal‑native agent workflow for scanning, repairing, and reviewing AI‑generated code.
Changes in this release
| Type | Severity | Summary | CVE |
|---|---|---|---|
| Feature | Medium |
Adds terminal-native agent workflow for scanning, repairing, and reviewing AI-generated code. Adds terminal-native agent workflow for scanning, repairing, and reviewing AI-generated code. Source: llm_adapter@2026-06-11 Confidence: high |
— |
| Feature | Medium |
Adds `aislop agent` command for provider-aware repair sessions from the terminal. Adds `aislop agent` command for provider-aware repair sessions from the terminal. Source: llm_adapter@2026-06-11 Confidence: high |
— |
| Feature | Medium |
Adds agent session, monitor, apply, watch, stop, and provider commands. Adds agent session, monitor, apply, watch, stop, and provider commands. Source: llm_adapter@2026-06-11 Confidence: high |
— |
| Feature | Medium |
Adds live session state showing edited files, score targets, token usage, and review steps. Adds live session state showing edited files, score targets, token usage, and review steps. Source: llm_adapter@2026-06-11 Confidence: high |
— |
| Feature | Medium |
Improves terminal output, command suggestions, scoped scans, and rule impact metadata. Improves terminal output, command suggestions, scoped scans, and rule impact metadata. Source: llm_adapter@2026-06-11 Confidence: high |
— |
| Feature | Medium |
Calibrates scoring to soften low‑impact style signals while keeping real findings visible. Calibrates scoring to soften low‑impact style signals while keeping real findings visible. Source: llm_adapter@2026-06-11 Confidence: high |
— |
| Performance | Medium |
Improves Python and Go dependency/import detection to reduce noisy false positives. Improves Python and Go dependency/import detection to reduce noisy false positives. Source: llm_adapter@2026-06-11 Confidence: high |
— |
| Bugfix | Medium |
Fixes agent worktree diffs to correctly include staged and untracked files. Fixes agent worktree diffs to correctly include staged and untracked files. Source: llm_adapter@2026-06-11 Confidence: high |
— |
| Bugfix | Medium |
Fixes monitor debounce handling so stable changes trigger the settled scan/repair cycle. Fixes monitor debounce handling so stable changes trigger the settled scan/repair cycle. Source: llm_adapter@2026-06-11 Confidence: high |
— |
| Bugfix | Medium |
Fixes telemetry coverage, subdirectory scope handling, monitor stop behavior, and installer stability. Fixes telemetry coverage, subdirectory scope handling, monitor stop behavior, and installer stability. Source: llm_adapter@2026-06-11 Confidence: high |
— |
| Bugfix | Low |
Ensures zero-debounce monitor mode (`--debounce 0`) reacts immediately to the first changed snapshot. Ensures zero-debounce monitor mode (`--debounce 0`) reacts immediately to the first changed snapshot. Source: granite4.1:30b@2026-06-11-audit Confidence: low |
— |
Full changelog
aislop v0.12.0 adds a terminal-native agent workflow for scanning, repairing, and reviewing AI-generated code without leaving your local checkout.
Highlights
- Added
aislop agentfor provider-aware repair sessions from the terminal. - Added agent session, monitor, apply, watch, stop, and provider commands.
- Added live session state with edited files, score targets, token usage, and review steps.
- Improved terminal output, command suggestions, scoped scans, and rule impact metadata.
- Calibrated scoring so low-impact style signals are softer while real findings stay visible.
- Improved Python and Go dependency/import detection to reduce noisy false positives.
Fixes
- Fixed agent worktree diffs so staged and untracked files are included correctly.
- Fixed monitor debounce handling so stable changes trigger the settled scan/repair cycle.
- Fixed telemetry coverage, subdirectory scope handling, monitor stop behavior, and installer stability.
- Fixed zero-debounce monitor mode so
--debounce 0reacts immediately to the first changed snapshot.
Validation
- 144 test files passing
- 1,276 tests passing
- Self-scan score: 100/100
- 0 diagnostics
npm pack --dry-runverified[email protected]with 26 files
Weekly OSS security release digest.
The CVE patches and breaking changes that affected production tools this week. One email, every Sunday.
No spam, unsubscribe anytime.
Share this release
About AISlop
All releases →Beta — feedback welcome: [email protected]