mirror of
https://github.com/Yamato-Security/WELA.git
synced 2026-11-18 18:16:06 +01:00
Assess native rule eligibility with explicit evidence gates
This commit is contained in:
1 parent
9d993a2a7e
commit
36ad97114c
17 files changed
+781
-27
No files matched your search
@@ -30,9 +30,9 @@ Each exact channel is read independently. An enabled AppLocker EXE/DLL channel d
|
||||
- Defender: WinDefend service state, runtime mode, antivirus/real-time/behavior/network inspection flags. Passive mode, ASR, network protection and controlled folder access have different prerequisites. An enabled channel is not proof of active protection or of all Defender events.
|
||||
- Other native providers: exact channel registration and available service state, including BITS, printing, WMI, Terminal Services, DFSN, Firewall and SMB client. Provider policy, activity and required event fields remain conditional. Application and System channels can receive events from many independent providers.
|
||||
|
||||
No rule receives current or ideal usable-rule credit merely because one of these channels is enabled. Their mapped rules remain in the complete, deduplicated corpus denominator and in `UnusableRules.csv` (meaning **not confirmed usable**, including conditional/unknown sources). A rule can still qualify through a separately assessed applicable source. The assessment does not claim a Sigma uplift or prove central ingestion. Existing advanced Security policy estimates retain their own SACL and other prerequisites.
|
||||
No rule receives current or ideal usable-rule credit merely because a channel or Security audit policy is enabled. Rules remain in the complete, deduplicated corpus inventory and in `UnusableRules.csv` (meaning **not confirmed usable**, including conditional/unknown sources). `RuleEligibility.csv` and JSON/HTML provide per-rule reasons and explicit native/full-corpus denominators. Configuration estimates remain separate from detection readiness. See [native rule eligibility](native-rule-eligibility.md) for the supported imported-evidence checks and their trust boundary.
|
||||
|
||||
This conservative change can lower the reported percentage. A future event-specific readiness model can promote individual native provider rules when all required policies, event types and fields have been validated; broad channel-level promotion would recreate the original problem.
|
||||
Without complete lab evidence, the reported Ready count is zero. This describes missing validation, not the usefulness of logging. The separate `rule-eligibility` command can review supported imported evidence for its recorded context/time; it never silently applies old lab results to the currently audited host. Additional native provider adapters still require reviewed policy, event, field and backend evidence.
|
||||
|
||||
## Validation and outstanding integration evidence
|
||||
|
||||
|
||||
Reference in new issue
Block a user