GitHub issue #100730 in anthropics/claude-code, opened 2026-10-09 by TheAviv, records a report that in an attended Claude Code cloud session the auto mode classifier refused a fire_trigger call on the account owner's own scheduled task, with the reason Cowork Run Routine Now, and refused or stopped a file transfer between two machines the owner controls as Data Exfiltration, and that there is no setting, prompt or override to approve the refused action. The author added further instances in comments through 2026-11-08. No maintainer response appears in the thread.
Verbatim Excerpts from Cited Records
“**Surface:** Claude Code cloud session (Cowork on claude.ai), attended by the account owner, model claude-opus-5-5, Max plan. Related: #97613 (auto mode classifier blocking work the user authorized, a regression since about 23 Sep) and #97308. This adds two concrete cases from 8 to 9 Oct 2026.”
·
anthropics/claude-code (issue 100730 opening body, Surface section)
·
2026-10-09T07:52:31Z
·GitHub Issue ↗
“- The agent called `mcp__claude-code-remote__fire_trigger` on that task with no arguments beyond a short context note. - The call was refused: `Permission for this action was denied by the Claude Code auto mode classifier. Reason: [Cowork Run Routine Now].`”
·
anthropics/claude-code (issue 100730 opening body, Case 1 section)
·
2026-10-09T07:52:31Z
·GitHub Issue ↗
“That tool exists to run my tasks on demand. Refusing it for the account owner, in an attended session, on a task the owner made, removes the feature without telling anyone. There is no setting, prompt or override to approve it.”
·
anthropics/claude-code (issue 100730 opening body, Case 1 section)
·
2026-10-09T07:52:31Z
·GitHub Issue ↗
“Both a local HTTP server in the sandbox, and later the equivalent on my own server serving my own browser, were refused or stopped as "[Data Exfiltration]".”
·
anthropics/claude-code (issue 100730 opening body, Case 2 section)
·
2026-10-09T07:52:31Z
·GitHub Issue ↗
Counter-record: Issue open; API returned 24 comments, all 24 read. All comments are by the author TheAviv, adding further instances of classifier refusals from 8 Oct through 8 Nov 2026, including a weekly rollup for 17 to 24 Oct 2026 and a new label on 8 Nov 2026. No maintainer response appears in the thread. No retraction of the report appears in the thread. The claim documents only the original report and the author's follow-up instances, not a verified current state.
Falsification probe: If issue 100730 is deleted without archive, if the author edits the opening body so the quoted sentences no longer match verbatim, or if the author posts a retraction of the report, this claim drops to Withdrawn.
Audit verification: Verified 100% continuous verbatim substring against official GitHub REST API by
scripts/verify_claim_verbatim.py.