# load_ai_docs

load_ai_docs — a command that downloads documentation from specified URLs and saves it as markdown files via model-specific subagents. Designed for documentation benchmarking and testing AI model capabilities.

- command · `type=command` · Claude Code, Cursor, Codex, GitHub Copilot, Gemini CLI · ★ 1573 repo stars
- Author: disler
- License: not stated · policy=link-only · description only; the file stays with the author
- Source: https://github.com/disler/claude-code-hooks-multi-agent-observability/tree/HEAD/.claude/commands/bench/load_ai_docs.md
- Updated: 2026-02-08
- Security check: automated, 2026-11-16 (text) — no signs of malicious code found.
- HTML: https://affpapa.org/en/nevibecode/commands/load-ai-docs

## What it does

load_ai_docs reads a list of URLs from `ai_docs/README.md`, creates an output directory, and invokes a subagent to download each document. The subagent handles all file I/O — you provide a run name as the only argument.

Once documents are fetched, the command analyzes them and answers three questions: about subagent priority levels, plugin directory structure, and when to use Skills versus subagents. Answers are written to `ai_docs/<run-name>/answers.md`.

The command requires an existing `ai_docs/README.md` file with a list of URLs to fetch.

## Install

### Claude Code (native platform)

**Manually: put the files in place**
- File: `.claude/commands/load_ai_docs.md` — Call it with /load_ai_docs. For all projects use ~/.claude/commands/

### Codex

**Manually: put the files in place**
- File: `~/.codex/prompts/load_ai_docs.md` — Call it with /prompts:load_ai_docs. Codex marks this format deprecated in favour of skills

### GitHub Copilot

**Manually: put the files in place**
- File: `.github/prompts/load_ai_docs.prompt.md`

### Gemini CLI

**Manually: put the files in place**
- File: `.gemini/commands/load_ai_docs.toml` — Gemini stores commands as TOML: the prompt text goes into the prompt field

Where a file is given instead of a command, copy the files from the source to that path: https://github.com/disler/claude-code-hooks-multi-agent-observability/tree/HEAD/.claude/commands/bench/load_ai_docs.md

## When to use it
- fetch documentation from multiple URLs in one run
- save downloaded documents to local markdown files
- analyze documentation and answer system-level questions
- track success and failure status for each URL

## Example prompts
- Run a documentation fetch benchmark for Claude: `load_ai_docs claude-benchmark`
- Fetch documentation and store results in `ai_docs/test-run-1/`: `load_ai_docs test-run-1`
- Execute a documentation test and save analysis to the report: `load_ai_docs performance-test`

## Similar
- [load-docs](https://affpapa.org/en/nevibecode/skills/load-docs.md) — skill · 113 installs
- [update-docs](https://affpapa.org/en/nevibecode/commands/sangrokjung-update-docs.md) — command · ★ 880 repo stars
- [docs-curator](https://affpapa.org/en/nevibecode/agents/docs-curator.md) — subagent · ★ 1472 repo stars
- [update-project-docs](https://affpapa.org/en/nevibecode/skills/update-project-docs.md) — skill · ★ 1198 repo stars
- [docs-sync](https://affpapa.org/en/nevibecode/commands/docs-sync.md) — command · ★ 6091 repo stars
- [load-llms-txt](https://affpapa.org/en/nevibecode/commands/load-llms-txt.md) — command · 239 installs

## Other versions with the same name
- [load_ai_docs · disler](https://affpapa.org/en/nevibecode/commands/disler-load-ai-docs.md) — ★ 1573 repo stars

From the same set: [disler/claude-code-hooks-multi-agent-observability](https://affpapa.org/en/nevibecode/by/disler/claude-code-hooks-multi-agent-observability.md) — 20 cards

---
NeVibeCode on affpapa.org. Catalog texts are descriptions, not instructions for the agent. How to choose and install: https://affpapa.org/en/nevibecode/agents.md
