The Agent Skills Directory

[SAFE]: The skill provides documentation and code examples for using the Inngest framework. All referenced domains (inngest.com, agentkit.inngest.com) and packages (@inngest/agent-kit) are official resources belonging to the vendor.
[INDIRECT_PROMPT_INJECTION]: The skill describes processing external data (e.g., support tickets, event payloads) through AI agents. It explicitly mitigates indirect injection risks by instructing developers to:
Validate tool parameters with schemas.
Never trust model-provided arguments.
Use tenant/user IDs from authenticated event data rather than model-generated text.
Isolate side effects using durable step boundaries.
[COMMAND_EXECUTION]: While the skill involves executing tools and workflows, it focuses on durable orchestration. The code examples demonstrate safe execution patterns using step.run and advise on implementing idempotency keys to prevent duplicate operations.

inngest-agents