From 94be0148ce91390ec8fffe155aa306d39b2a5735 Mon Sep 17 00:00:00 2001 From: dach-a Date: Wed, 18 Nov 2026 03:15:57 +0000 Subject: [PATCH] docs(security): document scope_distance as a trusted input States the trust boundary discussed in #317: - output.json's scope_distance is the integrated scanner's own verdict and is ingested as-is; in-scope events are not re-checked against ARTEX's rule set. - The enforced boundary is the task allow-list: assets reach the graph (insert_assets) only through EvaluateAssetGate, and rejects are recorded as in_scope_not_asset. - The no-allow-list case is an accepted gap, with operator guidance: configure an allow-list per task or review output.json before feeding it to ARTEX. Cross-checking ingested events against ARTEX's rule set would redefine existing scanner configurations, so it is a deliberate design question, not an accepted vulnerability. Option 2 as picked on #317; documentation only. --- SECURITY.md | 8 ++++++++ 1 file changed, 8 insertions(+) diff --git a/SECURITY.md b/SECURITY.md index 53091b7..a292dd2 100644 --- a/SECURITY.md +++ b/SECURITY.md @@ -35,6 +35,14 @@ ARTEX is an offensive-security tool. Test only against: Testing, scanning, or exploiting third-party systems without explicit authorization is strictly prohibited. Reports that violate this principle may be closed without action. +## Trusted Inputs and Scope Enforcement + +ARTEX trusts `output.json` as produced by its integrated scanner. The `scope_distance` field carried on scope-relevant events is the scanner's own verdict and is ingested as-is: an event the scanner marks in scope (per its own `-s`/`--no-scope` configuration) is treated as in scope without being re-checked against ARTEX's rule set. + +The enforced trust boundary is the **task allow-list**. Assets enter the asset graph (`insert_assets`) only through `EvaluateAssetGate`, which enforces the task's allow-list when one is configured; events rejected by the gate are recorded as `in_scope_not_asset` and cannot reach task execution. + +Accepted gap: when a task has no allow-list, the gate has nothing to enforce, and any host carried on an in-scope event enters the graph as an asset. Operators who rely on the scanner's scope rules alone should configure an allow-list per task, or review `output.json` before feeding it to ARTEX. Cross-checking ingested events against ARTEX's own rule set would change what existing scanner configurations mean for those operators, so it is treated as a deliberate design question (see #317), not an accepted vulnerability. + ## What to Expect This is an open-source project maintained on a best-effort basis. We aim to acknowledge a report within about a week and keep you updated, but we cannot guarantee specific response or fix timeframes.