Agents Sdk

What it does

The skill maps common agent application tasks to the Cloudflare Agents SDK's runtime concepts and documentation. It covers the relationship between server-side state, routed requests and connected clients, with additional references for scheduling and multi-step work. This helps an agent locate the right integration boundary before composing examples from unrelated frameworks.

Best for

Building or reviewing an application that already intends to use Cloudflare's agents package, especially when state persistence or ongoing client communication matters. Describe how the user starts a session, what state belongs to that session and which work continues after a request ends.

Tradeoffs and requirements

SDK features and experimental APIs can change. Persistent agent state does not by itself provide correct authorization or a safe tool-execution policy. Remote model providers and hosting resources may have separate costs even though this catalog's ingestion uses no paid AI service.

Unknown; inspect source requirements and supporting scripts.

Install from the source

Inspect the pinned source snapshot and its supporting files first. The command selects this skill through the open-source skills CLI. Node.js and agent-specific setup may be required. Installation can download code. Catalog ingestion never executes skill instructions or install scripts.

npx skills add https://github.com/cloudflare/skills/tree/83a27eb901c2c63f6caaf0eca6168993a322d268/skills/agents-sdk --skill agents-sdk

Review pinned skill folder ↗ · CLI documentation ↗

After source review, use npx skills check and npx skills update to inspect and apply upstream updates. Re-check changed instructions before running them.

Safety & review signals

Static review only. This is not a security audit or a guarantee of safety.

SKILL.md text only; scripts, dependencies and runtime behavior not audited.

• Network Access — review the source context.
• Shell Or Package Install — review the source context.

Allowed tools declaration: Unknown. Runtime permissions and network behavior: Unknown.

AI Craft Pad Skill Score

86/100 · Metadata coverage 100% · Model 1.0

Scores summarize observable discovery signals. Popularity and repository activity apply to the whole repository. Unknown runtime safety earns no certification.

Documentation

100/100

100

Maintenance

100/100

100

License Clarity

100/100

100

Repository Health

100/100

100

Compatibility Evidence

80/100

80

Popularity

63/100

63

Originality

80/100

80

Static Safety

60/100

60

Suggested workflow

Inspect the installed package, choose a narrow state and routing contract, implement one complete interaction and test its lifecycle. Pair with durable-objects when the underlying coordination and storage behavior needs closer review.

How to judge the result

Check state restoration, connection loss, method access controls and failed background work. Validate the deployed bindings and real client interaction rather than treating a successful bundle as completion.