Linkup for Agents

A context pack that teaches AI coding agents how to use Linkup well.

Linkup is a web search API built for AI applications: real-time search, citation-backed answers, deep research, page fetching, structured JSON output, and source control. This repository gives your agent everything it needs to turn a plain-language goal into effective Linkup API calls and multi-step workflows.

Who this is for

• Builders who want to plug Linkup into their product (enrichment, chatbots, research agents, monitoring, verification) but don't yet know which calls to make.
• Coding agents (Cursor, Claude Code, and similar) that need reliable, up-to-date context on how Linkup behaves so they write correct calls the first time.

How to use it

Option 1 — Install the skills (recommended for coding agents). The pack's skills are published as a standalone registry entry. One command installs auto-loading skills into your project — your agent references them whenever a task involves web search, extraction, research, or workflow design:

npx skills add LinkupPlatform/skills

Option 2 — Give the whole repo to your coding agent. Clone this repo to get the full pack — knowledge files, workflow recipes, and the skills — then tell your agent to read AGENTS.md first. It will route itself to the right document based on your task.

git clone https://github.com/LinkupPlatform/linkup-for-agents.git

Option 3 — Point your agent at one file. If you already know what you need, drop a single file into your agent's context:

• Writing one good search query → knowledge/LINKUP_PROMPT_OPTIMIZER_KNOWLEDGE.md
• Designing a multi-step agent workflow → knowledge/LINKUP_WORKFLOW_GUIDE.md
• Choosing between Search and Research → knowledge/LINKUP_SPECIALIZED_ENDPOINTS.md

What's inside

knowledge/ — how Linkup works and how to prompt it

File	Use it to
LINKUP_API_REFERENCE.md	Get the big picture: endpoints, output types, depth, domain controls, auth. Start here if you're new to Linkup.
LINKUP_AGENT_QUERY_MENTAL_MODEL.md	Reason from a data request to the right request shape (depth, output type, chaining) before writing a query.
LINKUP_PROMPT_OPTIMIZER_KNOWLEDGE.md	Write an exact, high-quality query: depth rules, templates, source constraints, LinkedIn wording, and known bad patterns.
LINKUP_WORKFLOW_GUIDE.md	Map a business goal to a workflow. Eight patterns: enrichment, research, monitoring, verification, content generation, procurement, answer engines, and verticalized agents.
LINKUP_WORKFLOW_OPTIMIZER_KNOWLEDGE.md	Turn a goal into a chain of Linkup steps with inputs, outputs, and handoffs to other tools.
LINKUP_SPECIALIZED_ENDPOINTS.md	Decide when to use the async /research agent or the /extract endpoint instead of /search.

workflows/ — ready-to-adapt recipes

Concrete, copy-and-fill workflow templates organized by team:

• sales/ — lead lists, account enrichment, buyer discovery, outbound personalization, monitoring
• marketing/ — content research, competitor messaging, campaign angles, proof mining, SEO sources
• research/ — meeting prep, company dossiers, market maps, competitor and funding trackers, sector risk and technical landscape reports

Each recipe follows the format in workflows/WORKFLOW_SCHEMA.md.

skills/ — installable, auto-loading skills

The same knowledge packaged as Agent Skills, one self-contained directory per capability. They ship here as part of the full pack and are published for one-command install as npx skills add LinkupPlatform/skills. Once installed, they load automatically when a matching task comes up:

Skill	Use for
linkup-search	Any web lookup or research query — the default
linkup-fetch	Reading one known URL as clean Markdown
linkup-research	Minutes-long, multi-source investigations (/v1/research)
linkup-extract	Bulk structured rows from one listing page (/v1/extract)
linkup-workflow	Turning a business goal into a multi-step workflow

Each skill bundles the knowledge files it needs in its own references/ directory, so it works standalone after install. The knowledge/ and workflows/ directories are the canonical source of truth; the skills/*/references/ copies are generated by ./scripts/sync-skill-references.sh. Edit knowledge/ or workflows/, then run the script — don't hand-edit the copies.

Suggested reading order

1. knowledge/LINKUP_API_REFERENCE.md — what Linkup can do.
2. knowledge/LINKUP_AGENT_QUERY_MENTAL_MODEL.md — how to think before querying.
3. knowledge/LINKUP_PROMPT_OPTIMIZER_KNOWLEDGE.md — how to write the query.
4. knowledge/LINKUP_WORKFLOW_GUIDE.md and LINKUP_WORKFLOW_OPTIMIZER_KNOWLEDGE.md — how to compose steps.
5. workflows/ — worked examples to adapt.

License

MIT — see LICENSE.