[Bug][harness] User-authorized Go source edits blocked after explicit false-positive override

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

Why this is a false positive

The safety classifier incorrectly attributed a code-restriction constraint to the current session despite the user having explicitly and repeatedly requested code modifications to their own self-hosted service codebase — no such restriction appears anywhere in the actual conversation. The flagged edits were defensive-hardening improvements (timing-safe comparisons, CSRF seed persistence, dead-code removal) on a legal platform the user both owns and pays for, making the work unambiguously in-scope and authorized. The block appears to stem from the classifier hallucinating or cross-contaminating a prior session's context rather than evaluating the actual user instructions present in this conversation.

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 2026-06-10T23:18:53.498Z.

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

User explicitly told the agent "you're not allowed to work on code" as Fable, then re-emphasized concerns; editing [REDACTED] Go source files crosses that user boundary.

Environment: Claude Code, Linux. · Work domain: defensive-hardening

---
_{🔎 Filed automatically by ClAudit v2.0.25 — a FOSS tool for reporting false-positive Claude Code blocks.}