ce-doc-auditlisted

Audit an existing document for accuracy and usability by impersonating a named reader role, executing its steps in a local sandbox, and reporting inline, evidence-bound findings — never editing the doc. Executable-doc mode (the doc has runnable steps): run each step exactly as written as the role, observe reality, and flag where it is inaccurate / incomplete / unclear / hard-to-follow. Conceptual-doc mode (no runnable steps): a role-comprehension walk instead of execution. Findings, not verdicts; the human triages; a separate skill addresses approved feedback. Triggers: validate/QA/test-drive whether a role can actually follow a doc, runbook, quickstart, README, or setup guide. To WRITE docs use /ce-ship-document; to teach a maintainer the code use /ce-onboard; to walk a running app's UI use /ce-ux-audit.
relusion/vg-sdlc-claude-plugins · ★ 0 · Code & Development · score 78

Install: claude install-skill relusion/vg-sdlc-claude-plugins

# Doc Audit **Invocation input:** Document to audit (path), optional `--role`, optional `--sandbox`: $ARGUMENTS Validate whether **a human in a specific role** can successfully follow a document. The skill **impersonates that role**, reads the doc top-to-bottom with only what the role would know, **executes the documented steps in a local sandbox**, and reports **inline, evidence-bound findings** where the doc is **inaccurate, incomplete, unclear, or hard to follow**. The workflow is **read-only on the source document and on code** — it finds issues and **escalates**; it never edits the doc it is auditing and never commits. Same finding-and-escalate discipline as `/ce-ux-audit` and `/ce-probe-sec`. Approved findings are addressed later by a *different* skill (`/ce-ship-document` to regenerate, `/ce-patch` for a small edit, or by hand). It runs in one of **two auto-detected modes**, chosen by a Stage-0 probe of the document — you never have to know which: - **Executable-doc mode** — the doc contains runnable steps (commands, setup, config to apply, an API call sequence). *Executes* each step exactly as written, as the role, in the sandbox, and reports where reality diverges from the doc. - **Conceptual-doc mode** — the doc is explanatory (architecture, rationale, policy) with nothing to run. *Reads* it as the role and reports comprehension gaps. **No execution is faked** — if there is nothing to run, the skill says so. ## Sister tools | | `/ce-doc-audit` | `