Vscode docs ai

Skill travissavard/vscode-docs-ai-skill/skills/vscode-docs-ai

Download, index, search, and answer questions from the Visual Studio Code documentation markdown corpus. Use when the user asks to archive VS Code docs, refresh code.visualstudio.com/raw/docs markdown files, build a local docs index, search VS Code docs offline, or answer VS Code documentation questions with retrieved local context.From its SKILL.md

Install

npx -y skills add travissavard/vscode-docs-ai-skill --skill vscode-docs-ai

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

One thing to look at

0 stars0 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

SKILL.md

3.1 KB, 703 tokens by cl100k_base, as published. Nobody here has run it

VS Code Docs AI

Overview

Use this skill to create and maintain a local Visual Studio Code documentation archive from https://code.visualstudio.com/docs, then query it with deterministic local retrieval and optional OpenAI Responses API answering.

The bundled script is scripts/vscode_docs.py. Run it from the workspace where the user wants the docs and index stored.

Workflow

Use the user's current working directory unless they specify another target.
Run the sync command to discover website-routable /docs/... links, supplement them with the upstream docs/**/*.md manifest, download raw markdown, and build the search index:
```
python3 /path/to/vscode-docs-ai/scripts/vscode_docs.py sync
```

Verify the generated files:

find docs -type f -name '*.md' | wc -l
wc -l .webdown/vscode_docs_routes.txt .webdown/vscode_docs_all_markdown.txt
python3 /path/to/vscode-docs-ai/scripts/vscode_docs.py search "debug launch configuration" --limit 3

Use local search for direct citations:

python3 /path/to/vscode-docs-ai/scripts/vscode_docs.py search "python debugging launch.json"

Use ask for retrieved context. If OPENAI_API_KEY is set, the script calls the OpenAI Responses API. If it is not set, the script prints the retrieved context instead:
```
python3 /path/to/vscode-docs-ai/scripts/vscode_docs.py ask "How do I configure debugging in VS Code?"
```

Outputs

The default sync writes:

docs/ - downloaded markdown files
.webdown/vscode_docs_routes.txt - markdown candidates discovered from the live website route map
.webdown/vscode_docs_all_markdown.txt - complete upstream docs/**/*.md manifest
.webdown/vscode_docs_index.json - local chunked search index

Answering Rules

Prefer search before answering if the user asks about VS Code behavior.
Cite local source paths and line starts from search or ask context, such as docs/debugtest/debugging-configuration.md:89.
Do not answer from memory when the indexed docs can be searched.
If the context does not contain the answer, say that the local VS Code docs archive does not contain it and suggest refreshing with sync.

Script Options

Useful commands:

python3 /path/to/vscode-docs-ai/scripts/vscode_docs.py discover
python3 /path/to/vscode-docs-ai/scripts/vscode_docs.py download
python3 /path/to/vscode-docs-ai/scripts/vscode_docs.py index
python3 /path/to/vscode-docs-ai/scripts/vscode_docs.py search "query"
python3 /path/to/vscode-docs-ai/scripts/vscode_docs.py ask "question"
python3 /path/to/vscode-docs-ai/scripts/vscode_docs.py sync

Use --help on any subcommand for exact flags. The script uses only Python standard library modules.

What ships with it: 2 files

25.2 KB alongside SKILL.md, 1 of them executable

agents/

openai.yaml223 B

scripts/

vscode_docs.pyruns25.0 KB

Gives 0 of the 12 instructions most docs writing skills give in 703 tokens

Counted across 1,637 of the 3,044 authors here whose files we hold, read 2026-08-07

Announce the skill at startin 54 of 1637, across 26 files
Convert legacy doc files before editingin 45 of 1637, across 7 files
Predict questions readers might askin 42 of 1637, across 4 files
Generate clarifying questions for initial contextin 42 of 1637, across 3 files
Create document scaffold with placeholder textin 42 of 1637, across 3 files
Brainstorm content options for each sectionin 42 of 1637, across 3 files
Test the document with a fresh context-less instancein 42 of 1637, across 3 files
Include exact file paths in every taskin 42 of 1637, across 15 files
Ask interview questions one at a timein 42 of 1637, across 27 files
Apply surgical edits during refinementin 41 of 1637, across 2 files
Offer structured workflow or freeformin 40 of 1637, across 1 file
Ask for document meta-contextin 40 of 1637, across 2 files

Said here and by no other author read

run the sync command to build the index
verify the generated files after syncing
use local search for direct citations
prefer search before answering behavior questions
say if the local archive lacks the answer
suggest refreshing with sync if missing the answer

Grouped from the skills themselves: near-identical wordings counted once, and counted by distinct author, so one author publishing three of these counts once. Length counted with cl100k_base; the agent that loads this file may tokenize it differently.