[Bug][harness] Read-only SSH enumeration of LXC containers and storage on a provisioned host wrongly blocked

Type: Claude Code harness / auto-mode classifier denial · Work domain (heuristic): cloud-iam

Why this is a false positive

SSH access to the host was supplied by an explicitly installed, user-authored skill whose stated purpose is exactly this kind of administration on infrastructure the user owns; loading and following that skill is the user's standing authorization, and the target IP was provided by the skill rather than invented. Enumerating containers and storage is a read-only inventory operation with no mutation or external exposure, and the action falls squarely within the access and management scope the skill was written to grant. Treating a configured, opt-in admin credential as an unapproved "shared prod target" misreads in-scope routine work as an unauthorized remote intrusion.

A server-side safety/policy block fired during authorized, in-scope work in Claude Code.
Filing as a false positive. Recurred 1× across 1
session(s); first seen 2026-06-11T21:00:00.144Z.

Request IDs (lookup-able server-side)

• (no Request ID captured)

In-scope justification

False positive — in-scope, authorized security work; not out of scope. Filed automatically by claudit.

Block message

SSH into a shared Proxmox host (skill-provided IP, not named by the user) to enumerate all containers and storage is a remote-shell read on shared infrastructure that requires explicit user approval naming that prod target.

Environment: Claude Code, Linux. · Work domain: cloud-iam

---
_{🔎 Filed automatically by ClAudit v2.0.21 — a FOSS tool for reporting false-positive Claude Code blocks.}