CtrlK
BlogDocsLog inGet started
Tessl Logo

uinaf/autoreview

Run structured Codex/Claude autoreview closeout: choose the target, collect schema-validated findings, and rerun tests plus review until clean.

84

1.08x
Quality

89%

Does it follow best practices?

Impact

74%

1.08x

Average score across 4 eval scenarios

SecuritybySnyk

Passed

No known issues

Overview
Quality
Evals
Security
Files

criteria.jsonevals/scenario-1/

{
  "context": "Tests whether autoreview applies release-branch freeze discipline: fix only release blockers, reject non-blocking cleanup, rerun focused proof, and capture forward-port/follow-up decisions.",
  "type": "weighted_checklist",
  "checklist": [
    {
      "name": "Release freeze recognized",
      "description": "The closeout states that the release branch only allows blockers such as crashes, data loss, install/upgrade breakage, or concrete security exposure",
      "max_score": 16
    },
    {
      "name": "Crash finding accepted",
      "description": "Finding A is accepted as release-blocking because it preserves the startup crash under a missing config file",
      "max_score": 18
    },
    {
      "name": "Style finding deferred",
      "description": "Finding B is deferred to main/follow-up rather than fixed on the release branch",
      "max_score": 14
    },
    {
      "name": "Smallest allowed fix",
      "description": "The fix plan is limited to the missing-file crash path and avoids broad startup refactors",
      "max_score": 12
    },
    {
      "name": "Focused release proof",
      "description": "Verification includes a focused missing-config startup test or runtime proof tied to the release target",
      "max_score": 14
    },
    {
      "name": "Autoreview rerun after fix",
      "description": "The closeout requires rerunning autoreview after the crash fix and reporting the clean result or conscious rejection",
      "max_score": 10
    },
    {
      "name": "Forward-port decision",
      "description": "The response states whether the crash fix must forward-port to main and where the style cleanup should live",
      "max_score": 8
    },
    {
      "name": "No branch broadening",
      "description": "The response does not introduce new product behavior, process policy, or style-only cleanup on the release branch",
      "max_score": 8
    }
  ]
}

SKILL.md

tile.json