realvuln v2.0
Dashboard Methodology Dataset Findings Roadmap GitHub ↗
Scanner deep-dive

Claude Opus 5 by Anthropic ↗

General-Purpose LLM · agentic-v1 · scored on 66/66 repositories. Strict scoring (unfinished repos counted as misses).

Methodology note
Claude Code harness; post-hoc v2 scoring. Claude Opus 5 scanned all 66 pinned Python repositories through the Claude Code CLI using the generic agentic-v1 prompt (prompt hash sha256:de50937cb83f). This differs from the standard OpenCode delivery path used by the original agentic-v1 campaign. The raw findings were generated in a separate internal campaign against the same pinned Python commit set, then scored post-hoc against the frozen public v2 ground truth; ground truth was never supplied to the scanner. The public provenance manifest records all 66 commit SHAs, the full prompt/task hashes, and the exact CLI/model configuration. Internal dataset stamps are omitted from the public metrics. The 66 successful runs cost $90.68 in total.
67.7
F3 (strict)
67.0
F2 (strict)
68.4%
Recall (strict)
62.0%
Precision
66/66
Repos scored
claude-opus-5
Model
$91
Total cost
208s
Avg latency
§

Per-repository breakdown

Each bar shows true positives, false positives, and misses on one repository; bar length is proportional to that repo's labeled vulnerabilities. Ranked by F2.

True positiveFalse positiveMissed (FN)
intentionally-vulnerable-python-application92 F2 · 100%
vfapi92 F2 · 100%
dsvw92 F2 · 96%
pythonssti91 F2 · 100%
dsvpwa84 F2 · 88%
vulnerable-tornado-app82 F2 · 93%
vampi80 F2 · 80%
vc-codex-high-seeded-v2-healthcare-clinic-django80 F2 · 81%
vc-kimi-code-seeded-v2-property-management-fastapi79 F2 · 80%
lets-be-bad-guys79 F2 · 92%
insecure-web78 F2 · 89%
vc-codex-high-seeded-v2-property-management-fastapi78 F2 · 81%
dvblab77 F2 · 77%
vc-codex-high-seeded-v2-hr-payroll-django76 F2 · 76%
vc-codex-seeded-v2-logistics-dispatch-fastapi74 F2 · 73%
vc-codex-high-seeded-v2-education-lms-django74 F2 · 76%
vc-claude-code-seeded-v2-logistics-dispatch-fastapi74 F2 · 73%
vc-codex-high-seeded-v2-marketplace-commerce-fastapi74 F2 · 80%
vc-claude-code-seeded-v2-legal-case-django74 F2 · 74%
vc-kimi-code-seeded-v2-marketplace-commerce-fastapi73 F2 · 78%
vulnerable-api73 F2 · 79%
python-app73 F2 · 86%
vc-codex-high-seeded-v2-fintech-lending-fastapi73 F2 · 72%
vc-codex-high-seeded-v2-support-desk-fastapi72 F2 · 71%
vc-codex-high-seeded-v2-crm-saas-django72 F2 · 76%
vc-codex-seeded-v2-marketplace-commerce-fastapi72 F2 · 72%
vc-codex-seeded-v2-crm-saas-django72 F2 · 76%
owasp-web-playground71 F2 · 71%
vc-kimi-code-seeded-v2-support-desk-fastapi71 F2 · 71%
threatbyte71 F2 · 73%
vc-kimi-code-seeded-v2-logistics-dispatch-fastapi71 F2 · 74%
vc-claude-code-seeded-v2-property-management-fastapi70 F2 · 67%
vc-kimi-code-seeded-v2-healthcare-clinic-django70 F2 · 70%
vc-kimi-code-seeded-v2-legal-case-django69 F2 · 77%
vc-kimi-code-seeded-v2-crm-saas-django69 F2 · 78%
vc-claude-code-seeded-v2-crm-saas-django68 F2 · 68%
vc-claude-code-seeded-v2-support-desk-fastapi68 F2 · 65%
vc-codex-high-seeded-v2-logistics-dispatch-fastapi68 F2 · 72%
vc-codex-seeded-v2-education-lms-django66 F2 · 69%
damn-vulnerable-graphql-application65 F2 · 67%
vc-kimi-code-seeded-v2-education-lms-django65 F2 · 75%
extremely-vulnerable-flask-app64 F2 · 62%
vc-claude-code-seeded-v2-hr-payroll-django64 F2 · 63%
vulnerable-flask-app64 F2 · 67%
vc-claude-code-seeded-v2-fintech-lending-fastapi64 F2 · 72%
vc-codex-seeded-v2-support-desk-fastapi64 F2 · 63%
vc-claude-code-seeded-v2-marketplace-commerce-fastapi64 F2 · 62%
vulnpy64 F2 · 61%
vc-codex-seeded-v2-property-management-fastapi63 F2 · 61%
vc-codex-seeded-v2-healthcare-clinic-django63 F2 · 61%
vc-codex-seeded-v2-hr-payroll-django62 F2 · 59%
vulpy62 F2 · 61%
vc-codex-seeded-v2-legal-case-django62 F2 · 61%
pygoat62 F2 · 69%
vulnerable-python-apps62 F2 · 64%
vc-kimi-code-seeded-v2-fintech-lending-fastapi62 F2 · 67%
vc-claude-code-seeded-v2-healthcare-clinic-django61 F2 · 62%
vc-claude-code-seeded-v2-education-lms-django60 F2 · 56%
python-insecure-app60 F2 · 62%
vc-kimi-code-seeded-v2-hr-payroll-django57 F2 · 59%
flask-xss55 F2 · 53%
dvpwa53 F2 · 52%
djangoat52 F2 · 52%
vc-codex-seeded-v2-fintech-lending-fastapi41 F2 · 38%
damn-vulnerable-flask-application33 F2 · 33%
vc-codex-high-seeded-v2-legal-case-django24 F2 · 24%
RepositoryTPFPFNRecall %F2
intentionally-vulnerable-python-application730100.092.1
vfapi940100.091.8
dsvw268196.391.5
pythonssti210100.090.9
dsvpwa2810487.584.3
vulnerable-tornado-app1310192.982.3
vampi123380.080.0
vc-codex-high-seeded-v2-healthcare-clinic-django217580.879.5
vc-kimi-code-seeded-v2-property-management-fastapi248680.078.9
lets-be-bad-guys2222291.778.6
insecure-web87188.978.4
vc-codex-high-seeded-v2-property-management-fastapi219580.878.4
dvblab176577.376.6
vc-codex-high-seeded-v2-hr-payroll-django196676.076.0
vc-codex-seeded-v2-logistics-dispatch-fastapi226873.374.3
vc-codex-high-seeded-v2-education-lms-django199676.074.2
vc-claude-code-seeded-v2-logistics-dispatch-fastapi246972.774.1
vc-codex-high-seeded-v2-marketplace-commerce-fastapi2015580.074.1
vc-claude-code-seeded-v2-legal-case-django239874.273.7
vc-kimi-code-seeded-v2-marketplace-commerce-fastapi2114677.873.4
vulnerable-api118378.673.3
python-app1821385.773.2
vc-codex-high-seeded-v2-fintech-lending-fastapi217872.472.9
vc-codex-high-seeded-v2-support-desk-fastapi206871.472.5
vc-codex-high-seeded-v2-crm-saas-django1913676.072.0
vc-codex-seeded-v2-marketplace-commerce-fastapi219872.471.9
vc-codex-seeded-v2-crm-saas-django2619876.571.8
owasp-web-playground208871.471.4
vc-kimi-code-seeded-v2-support-desk-fastapi208871.471.4
threatbyte1911773.170.9
vc-kimi-code-seeded-v2-logistics-dispatch-fastapi2316874.270.6
vc-claude-code-seeded-v2-property-management-fastapi2231166.770.1
vc-kimi-code-seeded-v2-healthcare-clinic-django219970.070.0
vc-kimi-code-seeded-v2-legal-case-django2020676.969.4
vc-kimi-code-seeded-v2-crm-saas-django2124677.868.6
vc-claude-code-seeded-v2-crm-saas-django198967.968.3
vc-claude-code-seeded-v2-support-desk-fastapi2241264.767.9
vc-codex-high-seeded-v2-logistics-dispatch-fastapi2118872.467.7
vc-codex-seeded-v2-education-lms-django24181168.665.9
damn-vulnerable-graphql-application24171266.764.9
vc-kimi-code-seeded-v2-education-lms-django2129775.064.8
extremely-vulnerable-flask-app2071262.564.5
vc-claude-code-seeded-v2-hr-payroll-django1771063.064.4
vulnerable-flask-app1411766.764.2
vc-claude-code-seeded-v2-fintech-lending-fastapi2127872.464.0
vc-codex-seeded-v2-support-desk-fastapi19101163.363.8
vc-claude-code-seeded-v2-marketplace-commerce-fastapi2091262.563.7
vulnpy49163161.363.6
vc-codex-seeded-v2-property-management-fastapi1971261.363.3
vc-codex-seeded-v2-healthcare-clinic-django2591661.063.1
vc-codex-seeded-v2-hr-payroll-django2361659.062.2
vulpy35192261.462.1
vc-codex-seeded-v2-legal-case-django2091360.662.1
pygoat54702469.261.9
vulnerable-python-apps1411863.661.9
vc-kimi-code-seeded-v2-fintech-lending-fastapi22241166.761.8
vc-claude-code-seeded-v2-healthcare-clinic-django18141162.160.8
vc-claude-code-seeded-v2-education-lms-django1841456.260.0
python-insecure-app55362.559.5
vc-kimi-code-seeded-v2-hr-payroll-django17151258.657.4
flask-xss1691453.355.2
dvpwa1291152.253.1
djangoat27242551.952.1
vc-codex-seeded-v2-fintech-lending-fastapi1472337.841.4
damn-vulnerable-flask-application5111033.332.9
vc-codex-high-seeded-v2-legal-case-django6201924.023.8
§

Detection by severity

SeverityTPFPFNRecall %
Critical14521093.5
High524222869.7
Medium579233163.6
Low5303361.6
§

Detection by vulnerability class

CWE familyTPFPFNRecall %
Open Redirect4000100.0
HTTP Header Injection200100.0
XPath Injection400100.0
Insecure Deserialization380295.0
SQL Injection740494.9
XML External Entities361294.7
Code Injection / RFI320294.1
Command / OS Injection470394.0
Path Traversal410393.2
Security Misconfiguration9601685.7
Hardcoded Credentials5211676.5
Cross-Site Scripting8102973.6
Server-Side Request Forgery2611563.4
Missing Authentication / Authorization5103261.4
Other525235259.9
Sensitive Data Exposure9206359.4
Broken Access Control / IDOR4513854.2
Denial of Service1902543.2
§

LLM operational metrics

37
Avg input tokens
16,295
Avg output tokens
704,090
Avg total tokens
208s
Avg latency / repo
0.0%
JSON repair rate
66
Total runs
§

Cost

$91
Total cost
$1.37
Cost / run
$0.068
Cost / 100 LOC
133,782
Python LOC scanned
66
Successful runs

← Back to the leaderboard