config.yaml Reference

Main configuration file at .jnkn/config.yaml.

Full Schema

# Jnkn Configuration

# Parsing settings
parsing:
  # File extensions to parse (auto-detected if not specified)
  extensions:
    python: [".py", ".pyi"]
    terraform: [".tf", ".tfvars"]
    kubernetes: [".yaml", ".yml"]
    javascript: [".js", ".ts", ".mjs"]

  # Python-specific settings
  python:
    # Additional extractors to load
    extra_extractors:
      - my_package.MyExtractor

# Stitching settings
stitching:
  # Minimum confidence for creating edges (0.0 - 1.0)
  min_confidence: 0.5

  # Token configuration
  min_token_length: 3

  # Tokens that provide no signal (ignored in matching)
  blocked_tokens:
    - id
    - key
    - url
    - host
    - port

  # Tokens that reduce confidence when matched alone
  low_value_tokens:
    - aws
    - prod
    - dev
    - main
    - test

  # Confidence calculation weights
  confidence:
    signals:
      exact_match: 1.0
      normalized_match: 0.9
      token_overlap_high: 0.85    # > 80% overlap
      token_overlap_medium: 0.7   # 50-80% overlap
      suffix_match: 0.6
      prefix_match: 0.6
      contains: 0.5
      single_token: 0.4

    penalties:
      short_token: 0.5      # Token < 4 chars
      common_token: 0.7     # Generic words
      ambiguous: 0.8        # Multiple possible matches
      low_value: 0.9        # Low-value tokens

  # Per-rule overrides
  rule_overrides:
    EnvVarToInfraRule:
      min_confidence: 0.6
      blocked_tokens:
        - host
    K8sToSecretRule:
      min_confidence: 0.4

  # Disable specific rules
  disabled_rules:
    - SomeUnwantedRule

  # Additional rules to load
  extra_rules:
    - my_package.MyRule

# Analysis settings
analysis:
  # Default max depth for blast radius (-1 = unlimited)
  default_max_depth: -1

  # Include transitive dependencies
  include_transitive: true

# Storage settings
storage:
  # Database path
  path: .jnkn/jnkn.db

  # Enable WAL mode for better concurrency
  wal_mode: true

# Output settings
output:
  # Default output format
  default_format: json

  # Include metadata in output
  include_metadata: true

  # Pretty print JSON
  pretty: true

# Logging
logging:
  level: INFO  # DEBUG, INFO, WARNING, ERROR
  format: "%(asctime)s - %(name)s - %(levelname)s - %(message)s"

Minimal Config

Most settings have sensible defaults. A minimal config:

stitching:
  min_confidence: 0.5

Section Reference

parsing

Controls how files are discovered and parsed.

Key	Type	Default	Description
extensions	dict	auto	File extensions per language
python.extra_extractors	list	[]	Additional Python extractors

stitching

Controls how cross-domain links are created.

Key	Type	Default	Description
min_confidence	float	0.5	Minimum confidence threshold
min_token_length	int	3	Minimum token length
blocked_tokens	list	[]	Tokens to ignore
low_value_tokens	list	[]	Tokens that reduce confidence
confidence.signals	dict	see above	Signal weights
confidence.penalties	dict	see above	Penalty multipliers
rule_overrides	dict	{}	Per-rule settings
disabled_rules	list	[]	Rules to disable
extra_rules	list	[]	Additional rules to load

analysis

Controls blast radius and other analysis.

Key	Type	Default	Description
default_max_depth	int	-1	Default traversal depth
include_transitive	bool	true	Include transitive deps

storage

Controls the database.

Key	Type	Default	Description
path	string	.jnkn/jnkn.db	Database file path
wal_mode	bool	true	Enable WAL mode

output

Controls output formatting.

Key	Type	Default	Description
default_format	string	json	Default output format
include_metadata	bool	true	Include extra metadata
pretty	bool	true	Pretty print JSON

logging

Controls logging.

Key	Type	Default	Description
level	string	INFO	Log level
format	string	see above	Log format string