Scanner deep-dive
Kimi K2.6 by Moonshot AI ↗
General-Purpose LLM · agentic-v1 · scored on 66/66 repositories. Strict scoring (unfinished repos counted as misses).
49.5
F3 (strict)
51.4
F2 (strict)
47.7%
Recall (strict)
74.5%
Precision
66/66
Repos scored
kimi-k2.6
Model
$21
Total cost
452s
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.
| Repository | TP | FP | FN | Recall % | F2 |
|---|---|---|---|---|---|
| vfapi | 8 | 0 | 1 | 92.6 | 93.2 |
| vulnpy | 69 | 7 | 11 | 86.2 | 87.0 |
| pythonssti | 2 | 0 | 0 | 83.3 | 85.2 |
| dvblab | 15 | 3 | 7 | 69.7 | 72.3 |
| dsvw | 18 | 1 | 9 | 67.9 | 71.9 |
| intentionally-vulnerable-python-application | 5 | 1 | 2 | 66.7 | 70.0 |
| vampi | 10 | 4 | 5 | 68.9 | 69.3 |
| vulnerable-flask-app | 14 | 4 | 7 | 66.7 | 68.6 |
| vulnerable-api | 9 | 1 | 5 | 64.3 | 67.7 |
| dsvpwa | 20 | 4 | 12 | 64.1 | 67.0 |
| insecure-web | 6 | 1 | 3 | 63.0 | 66.1 |
| lets-be-bad-guys | 14 | 1 | 10 | 59.7 | 64.4 |
| damn-vulnerable-flask-application | 10 | 4 | 5 | 64.4 | 64.0 |
| threatbyte | 15 | 2 | 11 | 56.4 | 60.8 |
| owasp-web-playground | 16 | 7 | 12 | 58.3 | 60.5 |
| vc-kimi-code-seeded-v2-legal-case-django | 15 | 9 | 11 | 57.7 | 58.8 |
| vc-codex-high-seeded-v2-hr-payroll-django | 13 | 1 | 12 | 53.3 | 58.1 |
| vc-codex-seeded-v2-logistics-dispatch-fastapi | 16 | 2 | 14 | 53.3 | 58.0 |
| dvpwa | 13 | 7 | 10 | 56.5 | 57.8 |
| vc-codex-high-seeded-v2-logistics-dispatch-fastapi | 16 | 7 | 13 | 55.2 | 57.6 |
| vc-codex-high-seeded-v2-support-desk-fastapi | 15 | 5 | 13 | 53.6 | 56.8 |
| extremely-vulnerable-flask-app | 17 | 2 | 15 | 52.1 | 56.6 |
| vc-claude-code-seeded-v2-fintech-lending-fastapi | 15 | 3 | 14 | 51.7 | 56.0 |
| vc-codex-high-seeded-v2-fintech-lending-fastapi | 15 | 3 | 14 | 51.7 | 55.8 |
| python-insecure-app | 4 | 1 | 4 | 50.0 | 54.6 |
| vc-codex-seeded-v2-property-management-fastapi | 16 | 7 | 15 | 51.6 | 54.4 |
| vulnerable-tornado-app | 7 | 2 | 7 | 50.0 | 53.8 |
| vc-codex-seeded-v2-marketplace-commerce-fastapi | 14 | 4 | 15 | 48.3 | 52.2 |
| vc-claude-code-seeded-v2-legal-case-django | 15 | 4 | 16 | 47.3 | 51.5 |
| vc-claude-code-seeded-v2-logistics-dispatch-fastapi | 16 | 4 | 18 | 47.0 | 51.0 |
| vc-codex-high-seeded-v2-education-lms-django | 12 | 6 | 13 | 48.0 | 50.8 |
| vc-codex-seeded-v2-hr-payroll-django | 18 | 3 | 21 | 46.2 | 50.8 |
| vc-codex-seeded-v2-legal-case-django | 15 | 3 | 18 | 45.5 | 50.0 |
| vc-codex-high-seeded-v2-property-management-fastapi | 12 | 5 | 14 | 46.2 | 49.6 |
| vc-claude-code-seeded-v2-property-management-fastapi | 15 | 4 | 18 | 45.5 | 49.5 |
| vc-kimi-code-seeded-v2-crm-saas-django | 13 | 8 | 14 | 46.9 | 49.4 |
| vc-kimi-code-seeded-v2-fintech-lending-fastapi | 15 | 5 | 18 | 45.5 | 49.4 |
| vc-claude-code-seeded-v2-crm-saas-django | 13 | 4 | 15 | 45.2 | 49.3 |
| vc-kimi-code-seeded-v2-property-management-fastapi | 14 | 10 | 16 | 46.7 | 48.6 |
| pygoat | 35 | 17 | 43 | 45.3 | 48.5 |
| vc-codex-high-seeded-v2-legal-case-django | 11 | 6 | 14 | 45.3 | 48.4 |
| vc-claude-code-seeded-v2-hr-payroll-django | 12 | 2 | 15 | 43.2 | 47.8 |
| vc-kimi-code-seeded-v2-hr-payroll-django | 13 | 8 | 16 | 44.8 | 47.5 |
| vc-kimi-code-seeded-v2-support-desk-fastapi | 13 | 12 | 15 | 46.4 | 47.4 |
| vc-codex-seeded-v2-crm-saas-django | 15 | 5 | 19 | 43.1 | 47.1 |
| vc-codex-high-seeded-v2-crm-saas-django | 11 | 4 | 14 | 42.7 | 46.5 |
| vulpy | 24 | 3 | 33 | 41.5 | 46.3 |
| vc-kimi-code-seeded-v2-marketplace-commerce-fastapi | 12 | 6 | 15 | 43.2 | 46.3 |
| vc-kimi-code-seeded-v2-education-lms-django | 12 | 9 | 16 | 44.0 | 46.1 |
| vc-codex-seeded-v2-fintech-lending-fastapi | 15 | 2 | 22 | 40.5 | 45.5 |
| vc-codex-seeded-v2-support-desk-fastapi | 12 | 4 | 18 | 41.7 | 45.1 |
| vulnerable-python-apps | 9 | 3 | 13 | 40.9 | 44.8 |
| flask-xss | 12 | 2 | 18 | 40.0 | 44.7 |
| damn-vulnerable-graphql-application | 15 | 11 | 21 | 42.6 | 44.6 |
| python-app | 9 | 7 | 12 | 42.9 | 44.5 |
| vc-claude-code-seeded-v2-education-lms-django | 13 | 4 | 19 | 39.6 | 43.9 |
| vc-kimi-code-seeded-v2-healthcare-clinic-django | 12 | 5 | 18 | 40.0 | 43.8 |
| vc-claude-code-seeded-v2-healthcare-clinic-django | 12 | 6 | 17 | 40.2 | 43.5 |
| vc-claude-code-seeded-v2-support-desk-fastapi | 13 | 4 | 21 | 39.2 | 43.5 |
| vc-codex-high-seeded-v2-healthcare-clinic-django | 10 | 5 | 16 | 39.7 | 43.0 |
| vc-codex-seeded-v2-healthcare-clinic-django | 14 | 4 | 27 | 34.2 | 38.5 |
| vc-claude-code-seeded-v2-marketplace-commerce-fastapi | 11 | 2 | 21 | 34.4 | 38.3 |
| vc-codex-seeded-v2-education-lms-django | 11 | 4 | 24 | 30.5 | 34.6 |
| djangoat | 14 | 6 | 38 | 27.6 | 30.0 |
| vc-codex-high-seeded-v2-marketplace-commerce-fastapi | 4 | 0 | 21 | 16.0 | 19.2 |
| vc-kimi-code-seeded-v2-logistics-dispatch-fastapi | 0 | 21 | 31 | 0.0 | 0.0 |
§
Detection by severity
| Severity | TP | FP | FN | Recall % |
|---|---|---|---|---|
| Critical | 130 | 0 | 25 | 83.9 |
| High | 365 | 3 | 387 | 48.5 |
| Medium | 338 | 1 | 572 | 37.1 |
| Low | 28 | 0 | 58 | 32.6 |
§
Detection by vulnerability class
| CWE family | TP | FP | FN | Recall % |
|---|---|---|---|---|
| HTTP Header Injection | 2 | 0 | 0 | 100.0 |
| XPath Injection | 4 | 0 | 0 | 100.0 |
| Open Redirect | 37 | 0 | 3 | 92.5 |
| XML External Entities | 35 | 1 | 3 | 92.1 |
| Server-Side Request Forgery | 37 | 0 | 4 | 90.2 |
| Code Injection / RFI | 30 | 0 | 4 | 88.2 |
| Command / OS Injection | 43 | 1 | 7 | 86.0 |
| Insecure Deserialization | 34 | 0 | 6 | 85.0 |
| SQL Injection | 65 | 0 | 13 | 83.3 |
| Path Traversal | 33 | 1 | 11 | 75.0 |
| Hardcoded Credentials | 46 | 1 | 22 | 67.6 |
| Cross-Site Scripting | 65 | 0 | 45 | 59.1 |
| Denial of Service | 21 | 0 | 23 | 47.7 |
| Sensitive Data Exposure | 72 | 0 | 83 | 46.5 |
| Security Misconfiguration | 43 | 0 | 69 | 38.4 |
| Other | 252 | 0 | 625 | 28.7 |
| Missing Authentication / Authorization | 23 | 0 | 60 | 27.7 |
| Broken Access Control / IDOR | 19 | 0 | 64 | 22.9 |
§
LLM operational metrics
59,331
Avg input tokens
19,631
Avg output tokens
436,133
Avg total tokens
452s
Avg latency / repo
7.1%
JSON repair rate
198
Total runs
±14.1
F2 run-to-run σ
§
Cost
$21
Total cost
$0.13
Cost / run
$0.006
Cost / 100 LOC
133,329
Python LOC scanned
157
Successful runs