Benchmark Quality Gate

feat: agent-audit v0.8.0 - Context-Aware Analysis & Tier Calibration #5

Sign in to view logs

Triggered via push February 5, 2026 09:23

HeadyZhang

master

Status Success

Total duration 36s

Artifacts 2

benchmark.yml

on: push

precision-recall

Agent-Vuln-Bench

Multi-Tool Comparison (Optional)

Quality Gate Check (v2)

Annotations

4 errors, 4 warnings, and 3 notices

Agent-Vuln-Bench

Process completed with exit code 1.

Agent-Vuln-Bench

Process completed with exit code 1.

precision-recall

Process completed with exit code 1.

Quality Gate Check (v2)

Process completed with exit code 1.

Agent-Vuln-Bench

No files were found with the provided path: results/baseline.json. No artifacts will be uploaded.

precision-recall

F1 82.71% below 0.87 target

precision-recall

Recall 80.08% below 85% target

precision-recall

Precision 85.52% below 90% target

fixture-count

Ground truth coverage good (81 samples)

fixture-count

Fixture count good (84)

atlas-coverage

ATLAS coverage good (32 rules)

Artifacts

Produced during runtime

Name	Size	Digest
avb-results Expired	3.35 KB	`sha256:7e17df988f886455b1039cc4b3ad2a79815e6b54ebd909c16eef27aca09c93ad`
layer1-results Expired	2.77 KB	`sha256:bbcbee667270e08369a8b4722d0c341388ed6550bc84c89e1afe73ab798c2a97`