{"data":{"id":"12c677f7-16dd-4996-87a0-54a63e6dc207","title":"Why the Harness Matters: Early Lessons From the Evo Code Scanning Agent Preview","summary":"Snyk's experimental Evo Code Scanning Agent, in preview with select customers, was compared against Claude Security and Codex Security using the same off-the-shelf models at equal compute budget. Across 14 applications and 1,426 known vulnerabilities, Evo found more known vulnerabilities than either lab tool (57% versus 32% for Claude Security, 66% versus 43% for Codex Security), but both lab tools achieved higher precision (94% and 97.5% versus 65% and 90%).","solution":"N/A -- no mitigation discussed in source.","labels":["industry","security"],"sourceUrl":"https://snyk.io/blog/why-harness-matters-evo-code-scanning-agent/","publishedAt":"2026-11-19T04:00:00.000Z","cveId":null,"cweIds":null,"cvssScore":null,"cvssSeverity":null,"severity":"info","attackType":[],"issueType":"news","affectedPackages":null,"affectedPackageNames":null,"affectedPackageRefs":null,"affectedVendors":["Anthropic","OpenAI"],"affectedVendorsRaw":["Evo Code Scanning Agent","Snyk","Claude Security","Codex Security","Claude Sonnet 5.5","GPT-6.1 Sol"],"classifierModel":"claude-haiku-5-5","classifierPromptVersion":"v4","summaryPromptVersion":"v2","headline":null,"headlinePromptVersion":null,"cvssVector":null,"attackVector":null,"attackComplexity":null,"privilegesRequired":null,"userInteraction":null,"exploitMaturity":null,"epssScore":null,"epssCheckedAt":null,"kevDateAdded":null,"advisoryAliases":null,"affectedPackagesSource":null,"affectedPackagesCheckedAt":null,"patchAvailable":null,"disclosureDate":"2026-11-19T04:00:00.000Z","fixReleasedAt":null,"daysToFix":null,"fixCheckedAt":null,"capecIds":null,"crossRefCount":0,"attackSophistication":"moderate","impactType":null,"aiComponentTargeted":"agent","llmSpecific":true,"classifierConfidence":0.8,"researchCategory":null,"atlasIds":null}}