code-walkthrough
Code Walkthrough
Generate a structured, linear walkthrough of a codebase (or a single file) using Showboat. The output is a markdown document meant to be read top-to-bottom in one sitting — a single narrative thread that builds understanding incrementally, not a reference manual you jump around in.
What "linear" means: The agent decides on a teaching sequence and the reader follows it from beginning to end. This is deliberately different from API docs (jump to what you need), READMEs (overview), code comments (scattered inline), or wiki pages (interlinked, no single path). A linear walkthrough has a beginning, middle, and end — like a guided tour where the guide chooses the route.
The walkthrough mixes narrative explanation with real code snippets extracted via shell commands — never from memory.
Why Showboat
The agent writes showboat exec demo.md bash "cat src/main.py | head -30" instead of pasting code from its context window. The document contains executed output, not recalled content. This eliminates hallucination risk entirely — every snippet is ground truth from the filesystem.
Showboat also supports showboat verify to re-run all code blocks and diff against recorded output — making walkthroughs reproducible proof of work, not just documentation.
Prerequisites
Showboat must be installed. Check with showboat --version. If missing:
uv tool install showboat
More from ianphil/my-skills
astral-uv
>
11glab
GitLab CLI (glab) for merge requests, issues, and CI/CD pipelines. Use when working with GitLab repositories for MR creation/review, issue management, pipeline debugging, or any GitLab API operations. Triggers on GitLab URLs, mentions of "merge request" or "MR" (not "PR"), gitlab.com, or glab commands.
10ado
Azure DevOps CLI (az ado). Use for work items, PRs, pipelines, and backlog management. Triggers on: az ado, ADO, azure devops, work item, backlog, az boards, az repos, az pipelines.
1ainotes
This skill should be used when the user asks to "consolidate notes", "summarize ainotes", "clean up notes", "ainotes", or wants to consolidate accumulated agent observations into a compact summary.
1workiq
This skill should be used when the user asks to "install WorkIQ", "set up WorkIQ", "query my emails with AI", "connect AI to Microsoft 365", "query my meetings or documents", "use workiq ask", or wants to use natural language to search Microsoft 365 data (emails, meetings, Teams messages, documents, people) from an AI assistant.
1closer
This skill should be used when the user asks to "close feature", "archive feature", "complete feature N", "move feature to completed", or wants to move a finished feature to the _completed/ directory.
1