cora-code 0.8.3

CLI-first AI code review — BYOK, diff/scan/branch, pre-commit hooks
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53

name: security-first
description: "Strict security focus — zero tolerance for vulnerabilities"
version: "1.0"

focus_areas:
  - id: security
    weight: 10
    action: block
    rules:
      - "No hardcoded secrets, credentials, or API keys in source code"
      - "Validate and sanitize all external inputs (HTTP params, file uploads, CLI args)"
      - "Use parameterized queries or ORM — never string interpolation for SQL"
      - "Check authorization on every endpoint and sensitive operation"
      - "Sanitize user-generated content before rendering (XSS prevention)"
      - "Use constant-time comparison for secrets and tokens"
      - "Never roll custom crypto — use established libraries"

  - id: injection
    weight: 10
    action: block
    rules:
      - "No SQL/NoSQL/XSS/SSRF/LDAP/Path traversal injection vectors"
      - "Escape all output in templates and HTML contexts"
      - "Validate file paths — reject path traversal patterns (../, ..\\)"
      - "Restrict SSRF — validate URLs before HTTP calls"

  - id: error_handling
    weight: 7
    action: warn
    rules:
      - "Never expose stack traces, internal paths, or system info to users"
      - "Log security events (auth failures, access denied) properly"
      - "Handle all error paths explicitly — no silent failures"

  - id: secrets_management
    weight: 8
    action: block
    rules:
      - "Load secrets from env vars or secret managers, not config files"
      - "No private keys, JWT secrets, or database passwords in code"
      - "Use HTTPS for all external communication"

ignore_areas:
  - style
  - naming
  - documentation
  - formatting

review_style:
  tone: strict
  detail_level: high
  suggest_fixes: true
  max_findings: null