[Question] Native offline command-resolution and effective-policy inspection in Claude Code 2.1.291

Status Closed — not planned
Reported on v2.1.291
Maintainer reply None cached
Activity 1 comment · opened Oct 9, 2026 · closed Oct 9, 2026

For Claude Code 2.1.291 on macOS, is there a supported native offline inspection command that, without starting a model request or accessing authentication/Keychain material, reports (1) the exact resolved source and effective expansion of a legacy project .claude/commands/pr-design.md command and (2) all effective configuration and managed-policy sources and precedence for a specified working directory and argv?

Please provide the exact command/flags and document startup filesystem writes, credential lookup, hooks, plugins, skills, memory loading and network effects, including MDM/server-managed policy. If there is no such operation in this version, please state that explicitly.

Standalone doctor is not sufficient merely because it is described as read-only: the installed doctor handler requests probeKeychain=true, and an installation/settings report alone does not establish custom-command resolution. Initialization-only and interactive status/skills menus also require an account of effects that occur before inspection. No repository files, logs, credentials or binary attachments accompany this question.

View original on GitHub ↗

This issue has 1 comment on GitHub. Read the full discussion on GitHub ↗