Skill Detail

Build event-driven document agent workflows with LlamaAgents

Use LlamaAgents to define async Python workflow steps, coordinate document-centric agent pipelines, and expose those workflows as services, clients, or deployable agent apps.

Templates & WorkflowsCustom Agents
Templates & Workflows Custom Agents Security Reviewed
⭐ 389 GitHub stars
COPY SKILL INSTRUCTIONS (OPTIONAL)
npx skills add agentskillexchange/skills --skill build-event-driven-document-agent-workflows-with-llamaagents Copy
Uses the third-party skills CLI, not an ASE-owned installer. Check your agent’s compatibility. This copies instructions; complete the upstream tool setup below separately.
At a glance
Tools required
Python; llama-index-workflows; optional llama-agents-server, llama-agents-client, and llamactl for serving and deployment
Install & setup
Install the workflow package with pip install llama-index-workflows; for deployable agent apps install llamactl with uv tool install llamactl, then initialize, serve, and deploy a bounded workflow.
Author
Run Llama
Publisher
Company
Last updated
Jun 6, 2026
Quick brief

Use this skill when an operator needs to turn messy document-agent work into a repeatable workflow: OCR, parsing, extraction, classification, validation, human review, and downstream handoff. The workflow is defined as async Python steps that emit and consume events, then it can run locally, be wrapped by `llama-agents-server`, called through `llama-agents-client`, or shipped with `llamactl`.

How it works

What this skill actually does

Invoke this instead of manual scripting when the document pipeline needs branching, loops, persistence, retries, streaming, or human review while staying inspectable as code. The scope boundary is custom Python document-agent orchestration with LlamaAgents packages and CLI. It is not a generic LlamaIndex listing, a broad multi-agent framework roundup, or a hosted product card.

Inputs and prerequisites: Python; llama-index-workflows; optional llama-agents-server, llama-agents-client, and llamactl for serving and deployment.

Setup notes: Install the workflow package with `pip install llama-index-workflows`; for deployable agent apps install `llamactl` with `uv tool install llamactl`, then initialize, serve, and deploy a bounded workflow.

Source and verification boundary: use https://developers.llamaindex.ai/python/llamaagents/ as the canonical reference before running the workflow; keep commands, API calls, CLI usage, and generated outputs reviewable against that upstream source.

Framework fit: publish this as a Custom Agents workflow only when the operator can invoke the documented toolchain directly, rather than treating the upstream project as a generic product listing.