heidi-anonymizer/example-config.yaml at main · stackabletech/heidi-anonymizer · GitHub

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
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
# Example configuration for Heidi Anonymizer
# This file demonstrates how to configure custom keywords and regex patterns
# for PII detection and redaction.

# Built-in recognizers to enable.
# If omitted, all five recognizers below are enabled with their default settings.
# Supported types: email, iban, credit_card, ip, spacy
recognizers:
  - type: email
  - type: iban
  - type: credit_card
  - type: ip
    ipv4_only: true # set to false to also detect IPv6 (may increase false positives)
  - type: spacy
    entities:
      - ORGANIZATION
      # - PERSON

# List of keywords that should be treated as PII if found
# These will be matched as whole words (deny-list based detection)
keywords:
  - "confidential"

# Custom regex patterns for detecting sensitive information
# Each pattern creates a new entity type based on the name
patterns:
  # Internal domain/hostname
  - name: "internal_domain"
    regex: "(?:^|\\s|:)([a-z0-9-]+\\.internal)(?:\\s|:|$)"
    score: 0.8
  - name: "cluster.local FQDN"
    regex: r"\b[a-z0-9]([-a-z0-9.]*[a-z0-9])?\.(svc|pod)\.cluster\.local\b"
    score: 0.9
  # Kubernetes namespaces specified as command line arguments
  - name: "-n <ns>"
    regex: r"(?<=-n )[a-z0-9][a-z0-9-]*"
    score: 0.7
  - name: "--namespace <ns>"
    regex: r"(?<=--namespace )[a-z0-9][a-z0-9-]*"
    score: 0.7
  - name: "--namespace=<ns>"
    regex: r"(?<=--namespace=)[a-z0-9][a-z0-9-]*"
    score: 0.7

# Language for NLP-based entity detection (default: "en")
language: "en"

# Optional NLP engine configuration.
# If omitted, defaults to spaCy with the en_core_web_lg model.
# For "spacy", model_name is a string.
# For "transformers", model_name must be a mapping with
# both "spacy" and "transformers" keys.

nlp_engine:
  nlp_engine_name: spacy
  models:
    - lang_code: en
      model_name: en_core_web_sm

# Example transformers NLP engine with spacy as "tokenizer".
#
# nlp_engine:
#   nlp_engine_name: transformers
#   models:
#     - lang_code: en
#       model_name:
#         spacy: en_core_web_sm
#         transformers: StanfordAIMI/stanford-deidentifier-base