# Claude Code

> Set up Spicrawl in Claude Code: add the hosted MCP server, install the agent skill, and add CLAUDE.md rules so the agent fetches web pages cheaply and handles errors correctly.

Source: https://docs.spicrawl.com/agents/claude-code

Three steps give Claude Code Spicrawl's tools, the knowledge to write Spicrawl code, and rules for using both well. Set your key first:

```bash
export SPICRAWL_API_KEY=spicrawl_live_...   # add to your shell profile
```

Or do all three with `spicrawl init --client claude`.

Or install the Spicrawl plugin, which bundles the MCP server and the skill, from the [Spicrawl/agent-plugins](https://github.com/Spicrawl/agent-plugins) marketplace:

```bash
claude plugin marketplace add Spicrawl/agent-plugins
claude plugin install spicrawl@spicrawl-plugins
```

Claude Code asks for your API key when the plugin is enabled and stores it as a sensitive value. With the plugin installed, skip steps 1 and 2.

## 1. Add the MCP server

For yourself, in this project:

```bash
claude mcp add --transport http spicrawl https://mcp.spicrawl.com/mcp \
  --header "Authorization: Bearer $SPICRAWL_API_KEY"
```

Your shell expands `$SPICRAWL_API_KEY` when you run the command, and Claude Code stores the header in `~/.claude.json`, outside the repository. Add `--scope user` to have the server in every project.

For the whole team, commit `.mcp.json` at the project root instead. Claude Code expands `${SPICRAWL_API_KEY}` from each person's environment when it starts:

```json title=".mcp.json"
{
  "mcpServers": {
    "spicrawl": {
      "type": "http",
      "url": "https://mcp.spicrawl.com/mcp",
      "headers": { "Authorization": "Bearer ${SPICRAWL_API_KEY}" }
    }
  }
}
```

The same command with `--scope project` writes this file for you, but with the expanded key in it; edit the header back to `${SPICRAWL_API_KEY}` before committing.

Check the connection with `claude mcp list`, or `/mcp` inside a session. The server shows 25 `spicrawl_*` tools. See [MCP server](https://docs.spicrawl.com/agents/mcp.md) for each tool.

## 2. Install the skill

```bash
mkdir -p .claude/skills/spicrawl
curl -fsSL https://app.spicrawl.com/skill.md -o .claude/skills/spicrawl/SKILL.md
```

Use `~/.claude/skills/spicrawl/` instead to have it in every project. Claude loads the skill when a task involves fetching or extracting web data, or when you type `/spicrawl`. See [Agent skill](https://docs.spicrawl.com/agents/skill.md).

## 3. Add rules to CLAUDE.md

Paste this section into `CLAUDE.md` at the project root. Claude Code reads it at the start of every session.

```markdown title="CLAUDE.md"
## Web data (Spicrawl)

Use Spicrawl to read web pages: the spicrawl_* MCP tools in chat, or
POST https://api.spicrawl.com/v1/scrape with `Authorization: Bearer $SPICRAWL_API_KEY` in code.
Docs: https://docs.spicrawl.com/llms.txt (append .md to any page URL for Markdown).

- Ask for markdown: `response_format: "markdown"` (API default is html); on spicrawl_scrape, `format: "markdown"`.
- Check the site's status, not only the HTTP status: `X-Target-Status` header, or `status`
  in the JSON envelope (spicrawl_scrape with `format: "json"`). 200 is the page; 404/410
  mean it does not exist; 403/429/503 mean the site refused, so escalate.
- On an error, switch on `code`. Retry only when `retryable` is true, after
  `retry_after_seconds`. Read `diagnostics.hint` and change what it names first.
  Never retry ERR::REQUEST::*, ERR::AUTH::* or ERR::LIMIT::QUOTA_EXCEEDED.
- Escalate one step at a time and stop at the first that works:
  plain fetch (1 credit) -> `js_render: true` (3) -> add the user's own `proxy`
  if they have one. Empty or skeleton content means render; ERR::UPSTREAM::CHALLENGE
  or a 403 target status means retry once, then the user's own proxy.
- Set `max_cost` on every request to the price of the step you intend.
- Keep the cache on (default). Set `cache: false` only for prices, stock or other live data.
- Trim tokens with `main_content_only` (on by default for markdown), `include_tags`, `exclude_tags`.
- For more than 20 URLs, use a batch job (spicrawl_batch_submit / POST /v1/batch).
  Batch items return raw HTML and apply only render, proxy and block_resources settings.
- Never print or commit SPICRAWL_API_KEY. Log `X-Request-Id` for failures.
```

## A first task to try

Start `claude` in the project and ask:

```text
Use Spicrawl to read https://example.com/pricing as markdown. List each plan with its monthly price.
If the page comes back empty, retry with rendering. Tell me how many credits it cost.
```

Claude calls `spicrawl_scrape`. To report the cost it needs `format: "json"`: the envelope carries `credits` and the site's `status`, while `format: "markdown"` returns only the document. For a coding task, try:

```text
Add a function fetch_page(url) to this project that calls the Spicrawl API for markdown,
following the Spicrawl skill's error-handling rules. Read the key from SPICRAWL_API_KEY.
```

Compare the result with the reference implementation in [Best practices](https://docs.spicrawl.com/agents/best-practices.md).

## Troubleshooting

| Symptom                                                                    | Fix                                                                                                                                                                                                   |
| -------------------------------------------------------------------------- | ----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
| `claude mcp list` shows spicrawl as failed, or 401                         | The key in the header is wrong, revoked, or the variable was empty. Re-run the `claude mcp add` command in a shell where `echo ${SPICRAWL_API_KEY:0:12}` prints `spicrawl_live_` or `spicrawl_test_`. |
| `.mcp.json` sends a literal `${SPICRAWL_API_KEY}`                          | Start `claude` from a shell where the variable is exported.                                                                                                                                           |
| Claude writes `js_render` on `spicrawl_scrape` and gets "unknown argument" | The tool's argument is `render`. The API field is `js_render`.                                                                                                                                        |
| Usage tools return `ERR::AUTH::INSUFFICIENT_SCOPE`                         | Grant the `read` scope to the key in the dashboard.                                                                                                                                                   |