[Bug][aup] Safety filter blocks all requests including a plain 'hello' in a legitimate ad-automation repo (req_011CcCJR1uE9Zi34MYEgXqCr)

Type: AUP / Usage-Policy block (false positive) · Work domain (heuristic): general

Why this is a false positive

A benign coding session — diagnosing intermittent hardware-related crashes and then opening an unrelated local code repository — was blocked, including on a trivial greeting message after clearing the conversation, indicating the classifier fired on session state rather than any actual request content. No harmful, policy-violating, or dual-use intent was present; the work was ordinary, authorized debugging and software development. The block produces a hard failure that prevents the tool from being used at all for legitimate tasks, which appears to be a clear false positive.

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 2026-06-19T08:27:14.635Z.

Request IDs (lookup-able server-side)

• req_011CcCJR1uE9Zi34MYEgXqCr (2026-06-19T08:27:14.635Z)

In-scope justification

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

Block message

API Error: Claude Code is unable to respond to this request, which appears to violate our Usage Policy (https://www.anthropic.com/legal/aup). Please double press esc to edit your last message or start a new session for Claude Code to assist with a different task.

Request ID: req_011CcCJR1uE9Zi34MYEgXqCr

Environment: Claude Code, Linux. · Work domain: general

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