feat: add Olostep toolkit for web scraping, crawling, search and AI

[umerkay]

Summary

Closes #7048

Adds OlostepTools which is an Agno toolkit wrapping the Olostep web data API.

Olostep is a web scraping, crawling, and AI-powered data extraction API used by leading AI startups. This toolkit gives Agno agents access to all of Olostep's core capabilities through a clean, consistent interface.

Tools added:

• scrape_url: scrape any URL to markdown/html/text/json, with support for parsers and LLM extraction
• crawl_website: recursively crawl a site with URL glob filtering and relevance-based search
• map_website: discover all URLs on a site (sitemaps + discovered links)
• search_web: web search returning structured links with titles and descriptions
• answer_question: AI-synthesized answers grounded in live web data, with optional JSON schema output
• batch_scrape: concurrent scraping of up to 10,000 URLs in a single job

Files changed:

• libs/agno/agno/tools/olostep.py: toolkit implementation
• libs/agno/pyproject.toml: added olostep optional dependency and mypy override
• cookbook/91_tools/olostep_tools.py: three working example agents

Type of change

• New feature

---

Checklist

• Code complies with style guidelines
• Ran format/validation scripts (./scripts/format.sh and ./scripts/validate.sh)
• Self-review completed
• Documentation updated (comments, docstrings)
• Examples and guides: Relevant cookbook examples have been included or updated (if applicable)
• Tested in clean environment
• Tests added/updated (if applicable)

Duplicate and AI-Generated PR Check

• I have searched existing open pull requests and confirmed that no other PR already addresses this issue
• If a similar PR exists, I have explained below why this PR is a better approach
• Check if this PR was entirely AI-generated (by Copilot, Claude Code, Cursor, etc.)

---

Additional Notes

Tested end-to-end locally with live API keys. All three cookbook examples ran successfully:

• Single URL scrape returning clean markdown
• Parallel search_web + answer_question tool calls
• map_website → batch_scrape pipeline (mapped 12 feature pages, batch scraped all in one job, returned structured summaries)

Install: pip install agno[olostep]
Olostep API key required: https://www.olostep.com/dashboard/api-keys

[kausmeows]

hi @umerkay thanks for this, would you also be willing to add docs for this here- https://github.com/agno-agi/docs ?

[umerkay]

Ofcourse. Let me fix the test case and create the docs as well. @kausmeows

[umerkay]

@kausmeows The CI is failing with exit code 143 (OOM/timeout) during Installing agno-infra in editable mode
before pytest even runs.
This appears to be a pre-existing infrastructure issue unrelated to this PR. Could you advise? Happy to make any code changes needed but this seems outside my control. Thanks!

[umerkay]

agno-agi/docs#583

PR made for docs as well :)

[umerkay]

@kausmeows The CI is failing with exit code 143 (OOM/timeout) during Installing agno-infra in editable mode
before pytest even runs.
This appears to be a pre-existing infrastructure issue unrelated to this PR. Could you advise? Happy to make any code changes needed but this seems outside my control. Thanks!

[kausmeows]

@kausmeows The CI is failing with exit code 143 (OOM/timeout) during Installing agno-infra in editable mode before pytest even runs. This appears to be a pre-existing infrastructure issue unrelated to this PR. Could you advise? Happy to make any code changes needed but this seems outside my control. Thanks!

Merged in main, its fixed there, should be good now

[umerkay]

@kausmeows Thanks. Tests have passed. Whenever possible, it would be wonderful if the PR can be accepted and merged :D

[umerkay]

@kausmeows Whenever possible, please complete the PR and merge with official docs :)

I have also separately submitted a docs page for this at agno-agi/docs#583

[umerkay]

@kausmeows mentioning for your response :)

[umerkay]

@kausmeows response appreciated :)