What probably matters to youSection of the release
Unclear What the optional path after prompt-audit changes is not stated.
What
/doctor now accepts prompt-audit, optionally followed by a path. With it, /doctor hands off to the prompt audit in the bundled claude-api skill. A skill is a packaged set of instructions Claude Code can use. The audit covers your project's instruction files:
CLAUDE.md and AGENTS.md
rules, skills, commands and agents
output styles
It is told not to read settings files, .mcp.json or ~/.claude.json, because those are not instructions and can hold secrets. If the claude-api skill is unavailable, the model explains why and stops. It can be unavailable because it is disabled, set to off in skillOverrides, or turned off with all bundled skills by disableBundledSkills or CLAUDE_CODE_DISABLE_BUNDLED_SKILLS. Before this, /doctor only took free-form extra instructions.
Why
It gives you an easy way to have the instruction files Claude reads reviewed, without exposing configuration files that may hold secrets.
Read from
Names in the bundle/doctor prompt-audit
Since it was published
The entry above is what we published on the day. These lines were added later, as Anthropic's own pages caught up, and they sit beside the original rather than replacing it.
Confirmed sinceAnthropic's documentation has since written up /doctor prompt-audit, on How Claude remembers your project.To have Claude check your instruction files for outdated or conflicting content, run `/doctor prompt-audit` in a session. Claude looks for problems such as instructions written for older models, references to files or commands that don't e…memorysee the edit
Added sinceA small documentation edit on How Claude remembers your project touched a line naming /doctor prompt-audit after this was published.To have Claude check these files for outdated or conflicting instructions, run `/doctor prompt-audit` in a session. Claude reads your CLAUDE.md, CLAUDE.local.md, and AGENTS.md files, plus the rules, skills, commands, subagents, and output …memorysee the edit
How sure we are
Something disagreesSomething we can check disagrees with this entry, or the writer said they could not settle it.
The writer flagged doubtWhat the optional path after `prompt-audit` changes is not stated.
Anthropic's documentation agreesA small documentation edit on How Claude remembers your project touched a line naming /doctor prompt-audit after this was published.
Anthropic's release notes agreeAdded /doctor prompt-audit (also /checkup prompt-audit) to audit your CLAUDE.md files, skills, agents and commands for prompting patterns…
