CalvinBackup/gstack
1
0
Fork 0
mirror of https://github.com/garrytan/gstack.git synced 2026-08-08 23:36:06 +02:00
Code Issues Packages Projects Releases Wiki Activity

feat: auto-scaled adversarial review (v0.9.5.0) (#297)

Browse Source 
* feat: auto-scaled adversarial review by diff size

Replace config-driven Codex review step with automatic adversarial review
that scales by diff size: small (<50 lines) skips adversarial, medium
(50-199) gets cross-model adversarial, large (200+) gets all 4 passes.

Adds Claude adversarial subagent as fallback when Codex unavailable.
Review log uses new "adversarial-review" skill name with source/tier fields.
Dashboard updated to read both adversarial-review and legacy codex-review.

* chore: regenerate SKILL.md files for auto-scaled adversarial review

* chore: bump version and changelog (v0.9.5.0)

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

* feat: allow user override of adversarial review tier

Users can now say "paranoid review", "run all passes", "full adversarial",
etc. to force the large tier regardless of diff size.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

---------

Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>
This commit is contained in:
Garry Tan
2026-03-21 12:12:56 -07:00
committed by GitHub
co-authored by Claude Opus 4.6
parent f075cb757f
commit 6c69febf11
15 changed files with 398 additions and 329 deletions
Download Patch File Download Diff File Expand all files Collapse all files
CHANGELOG.md
+8
View File
@@ -1,5 +1,13 @@
# Changelog
## [0.9.6.0] - 2026-03-21 — Auto-Scaled Adversarial Review
### Changed
- **Review thoroughness now scales automatically with diff size.** Small diffs (<50 lines) skip adversarial review entirely — no wasted time on typo fixes. Medium diffs (50–199 lines) get a cross-model adversarial challenge from Codex (or a Claude adversarial subagent if Codex isn't installed). Large diffs (200+ lines) get all four passes: Claude structured, Codex structured review with pass/fail gate, Claude adversarial subagent, and Codex adversarial challenge. No configuration needed — it just works.
- **Claude now has an adversarial mode.** A fresh Claude subagent with no checklist bias reviews your code like an attacker — finding edge cases, race conditions, security holes, and silent data corruption that the structured review might miss. Findings are classified as FIXABLE (auto-fixed) or INVESTIGATE (your call).
- **Review dashboard shows "Adversarial" instead of "Codex Review."** The dashboard row reflects the new multi-model reality — it tracks whichever adversarial passes actually ran, not just Codex.
## [0.9.5.0] - 2026-03-21 — Builder Ethos
### Added