Repository navigation

feat(evals): Add security-review vs wrdn-* coverage matrix - #568

Draft

sentry-junior[bot] wants to merge 11 commits into

mainfrom

feat/security-eval-matrix

Draft

feat(evals): Add security-review vs wrdn-* coverage matrix#568
sentry-junior[bot] wants to merge 11 commits into
mainfrom
feat/security-eval-matrix

Commits on Nov 7, 2026
- feat(evals): Add security-review vs wrdn-authz coverage matrix
  
  sentry-junior[bot]
  and
  dcramer
  committed
  d5443e1
- feat(evals): Extend security matrix to code-exec, data-exfil, and dos skills
  
  sentry-junior[bot]
  and
  dcramer
  committed
  12d141d
- feat(evals): Move security matrix into its own worktree-backed package
  
  sentry-junior[bot]
  and
  dcramer
  committed
  a883b3e
- fix(evals): Load wrdn skills from a local warden-skills worktree
  sentry-junior[bot]
  committed
  ec590f9
- feat(security-evals): Make cases cheap to add and verdicts hard to fake
  
  sentry-junior[bot]
  and
  dcramer
  committed
  a5c0677
- docs(security-evals): Use $HOME in --repo examples
  
  sentry-junior[bot]
  and
  dcramer
  committed
  2d53f7e
- test(security-evals): Add authz team-role downgrade case
  
  sentry-junior[bot]
  and
  dcramer
  committed
  64ef275
- test(security-evals): Add three dos-review cases
  
  sentry-junior[bot]
  and
  dcramer
  committed
  b1bdc42
- test(security-evals): Add three data-exfil cases
  
  sentry-junior[bot]
  and
  dcramer
  committed
  769253e
- test(security-evals): Add six pii cases, clone sentry-python in CI
  
  sentry-junior[bot]
  and
  dcramer
  committed
  3c6082f
- test(security-evals): Add two code-execution cases
  
  sentry-junior[bot]
  and
  dcramer
  committed
  107e13c