agentleFS
GeoDaoyuGeoDaoyu/arcgis-maps-sdk-js-ai-context/skills/arcgis-ai-components/AGENTS.md

arcgis-maps-sdk-js-ai-context / arcgis-ai-components

AGENTS.md0 starsChanged 7 months agoMITView on GitHub

Quick-reference decisions, checklists, and prerequisite verification for adding AI assistant capabilities. Before using AI components, verify: Recommendation: Include all three built-in agents for most applications. The assistant's orchestration automatically routes queries to the appropriate agent. Use built-in agents only when: Use custom agents (arcgis-assistant-agent) when:

The file

We split this AGENTS.md into the rules it gives an agent

4 rules. About 0 are common, 3 are distinctive, and 1 is very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈3

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 1

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Before using AI components, verify:

    line 7
  2. Distinctive:

    Use built-in agents only when:

    line 30
  3. Distinctive:

    No domain-specific logic is required

    line 34
  4. Very specific:

    Use custom agents (arcgis-assistant-agent) when:

    line 36

≈ “Common” is an estimate: at least 1 in 5 of the 40,246 AGENTS.md files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
AGENTS.md

Agent Guide: arcgis-ai-components

Quick-reference decisions, checklists, and prerequisite verification for adding AI assistant capabilities.

Prerequisites Checklist

Before using AI components, verify:

  • Web map has good layer metadata (descriptions, field aliases, tags)
  • Web map has embeddings stored as an item resource (see Web map setup guide)
  • End user is a signed-in named ArcGIS Online user (not trial, not public account)
  • End user has access to the web map and its layers
  • AI assistants are enabled in the ArcGIS Online organization settings
  • Beta apps are not blocked in the organization settings
  • End user has the role privilege to use AI assistants

Agent Selection Guide

User NeedAgent ComponentExample Prompts
Navigate the maparcgis-assistant-navigation-agent"Zoom to California", "Go to the largest feature"
Analyze dataarcgis-assistant-data-exploration-agent"What is the average?", "Show top 10 by value"
Understand the maparcgis-assistant-help-agent"What layers are here?", "Describe the data"
Domain-specific tasksarcgis-assistant-agent (custom)Depends on your tool definitions

Recommendation: Include all three built-in agents for most applications. The assistant's orchestration automatically routes queries to the appropriate agent.

Decision: Custom Agent vs Built-in Agents

Use built-in agents only when:

  • Your map has standard GIS data (feature layers with attributes)
  • Users need navigation, data queries, and map help
  • No domain-specific logic is required

Use custom agents (arcgis-assistant-agent) when:

  • You need integration with external APIs (weather, census, etc.)
  • Domain-specific business logic is required
  • You want specialized data processing or formatting
  • The built-in agents don't cover your use case

Layout Decision

LayoutWhen to Use
Side panel (calcite-shell-panel)Persistent assistant always visible alongside map
Expand widget (arcgis-expand)Space-constrained UI, assistant toggled on demand
Custom panelFramework-managed layout (React, Angular, etc.)

Troubleshooting

SymptomLikely CauseFix
No response to messagesNo agents registeredAdd agent components as children
Authentication errorUser not signed in or lacks privilegesCheck org settings and user roles
Agent misunderstands dataPoor layer metadataAdd field aliases, descriptions, embeddings
Agent can't find featuresMissing reference-elementSet to valid arcgis-map CSS selector

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.