.gitleaks.toml

# Repository-hygiene scan config (#151).
#
# Extends the upstream gitleaks default ruleset; adds NO custom rules.
# What it catches is documented at the level of the upstream tool's docs:
# https://github.com/gitleaks/gitleaks#rules
#
# The allowlist removes two structural false-positive classes this repo
# carries. (1) Bibliographic citation keys (e.g. "Becht2019UMAP",
# "vaswani2017-arxiv-v1") stored in key-shaped fields of eval gold tuples,
# the version-family example/design doc, and their schema test. These match
# the generic-api-key heuristic but are literature identifiers, not
# credentials — the 2026-06-10 baseline evaluation found 43 findings across
# 518 commits, all of this class, zero true positives. (2) Model-generated
# review prose in the frozen E4 evidence bundles (first hit 2026-08-05: a
# journal-tier descriptor in a promoted field_analysis.md). The bundles are
# byte-frozen promotion evidence (check_e4_promotion.py) that cannot be
# edited, and squash-merge makes per-commit fingerprints unpinnable on main,
# so the prose (.md) files are a rule-targeted allowlist entry below; machine
# logs and JSON records in the same bundles stay scanned, and every rule other
# than the misfiring generic-api-key stays live on the prose files too.
#
# Maintenance note: a new citation-tuple file under the allowlisted evals
# path needs no action here. A NEW false-positive class elsewhere should be
# added as a path entry (or a .gitleaksignore fingerprint for a one-off),
# never as a rule edit.

[extend]
useDefault = true

# Class 1: whole-file exemption (these files are key-shaped data by design).
[[allowlists]]
description = "Bibliographic citation keys misread as generic-api-key (literature identifiers, not credentials)"
paths = [
  '''evals/gold/citation_extraction/tuples/.*\.json''',
  '''scripts/fixtures/check_evals_gold_set/.*''',
  '''scripts/adapters/examples/.*''',
  '''scripts/adapters/tests/test_zotero\.py''',
  '''academic-paper/examples/version_family_reconciliation_example\.md''',
  '''docs/design/2026-05-28-kong-258-version-family-reconciliation\.md''',
  '''docs/design/2026-04-23-ars-v3\.6\.4-literature-corpus-adapters-plan\.md''',
  '''scripts/test_version_records_schema\.py''',
]

# Class 2 (rule-targeted): only the misfiring rule is exempted over the frozen
# E4 evidence prose; private-key / AWS / GitHub-PAT / Slack / JWT detection
# stays live over the same files.
[[allowlists]]
description = "Frozen E4 evidence-bundle model prose misread as generic-api-key (venue descriptors, not credentials)"
targetRules = ["generic-api-key"]
paths = ['''evals/heldout/reviewer_seeded_defects/runs/raw/.*\.md''']

# Class 3 (rule-targeted): #660's repository-owned synthetic conformance
# fixtures use citation_key values such as fixture_missing_2026. These are
# deterministic literature-record identifiers, not credentials. All other
# secret rules remain active over the same fixture bytes.
[[allowlists]]
description = "Synthetic tortured-phrase fixture citation keys misread as generic-api-key"
targetRules = ["generic-api-key"]
paths = [
  '''scripts/fixtures/tortured_phrase_screening/.*''',
  '''scripts/fixtures/bibliographic_integrity_signals/tortured_phrase_v1_2_.*\.json''',
]