Schema quality: 20 required parameters missing descriptions, 92 optional parameters undescribed

Running static schema analysis on the GitHub MCP server reveals 112 issues (20 errors, 92 warnings) across 26 tools. The most impactful: required parameters with no description, which forces AI agents to guess what values to provide.

**Summary:**

| Severity | Count | Impact |
|----------|------:|--------|
| Error (E103) | 20 | Required parameters with no description |
| Warning (W102) | 72 | Optional parameters with no description |
| Warning (W103) | 20 | Required string params with no enum/pattern/example |

**Examples of what agents see:**

`create_issue` requires `title` and `body` with no description. An agent doesn't know:
- Is `body` plain text or markdown?
- What's the maximum length?
- Should `title` include a prefix like `[BUG]`?

`search_repositories` requires `query` with no description. An agent doesn't know:
- Is this GitHub search syntax (`language:go stars:>100`)?
- Or plain text?
- What qualifiers are available?

`create_pull_request` requires `title`, `body`, `head`, `base` with no descriptions. An agent must guess:
- Is `head` a branch name or a ref like `refs/heads/feature`?
- Is `base` required to be `main` or can it be any branch?

**What good parameter descriptions look like:**

```json
{
"name": "query",
"type": "string",
"description": "GitHub search query using search syntax (e.g. 'language:go stars:>100'). See https://docs.github.com/en/search-github/searching-on-github"
}
```

**How this was found:**

```bash
mcp-assert lint --server "npx -y @modelcontextprotocol/server-github"
```

Adding descriptions to required parameters would significantly reduce agent misuse of these tools.