Skip to content
Back to skills

Data Sql Graph Agents

ASecurity

"Use Langroid structured-data agents for table/CSV/Pandas, SQL,

  • 247 stars
  • 0 votes
  • 0 copies
  • 4 views
  • Added September 8, 2026
ai-agentsrustgosqlapidatabasesecurity

Works with

  • api

Security analysis

A100/100

Pro scans all 7 files and shows the line behind each finding

Scanned September 8, 2026

npx -y skills add VectorSpaceLab/AREX-Skill --skill data-sql-graph-agents --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Data Sql Graph Agents?

Add the live security badge to your README. It updates with every re-scan.

Security grade badge for Data Sql Graph Agents
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/vectorspacelab-data-sql-graph-agents/badge)](https://www.skillsdirectory.com/skills/vectorspacelab-data-sql-graph-agents)

More formats (shields.io, HTML) on the badges page. Keep it an A: scan every change in CI with Pro.

Download with Pro
SKILL.md
---
name: data-sql-graph-agents
description: "Use Langroid structured-data agents for table/CSV/Pandas, SQL,
  Neo4j, ArangoDB, and CSV-to-knowledge-graph workflows with schema/context
  setup and safe query validation."
disable-model-invocation: true
metadata:
  disco-role: operating
license: MIT
---

# Data, SQL, and Graph Agents

Use this sub-skill for structured data questions, tabular analysis, SQL-backed
Q&A, and graph construction or graph Q&A.

## Choose the right agent

- Table, CSV, or Pandas analysis -> TableChatAgent
- SQL database Q&A -> SQLChatAgent
- Neo4j graph Q&A or writes -> Neo4jChatAgent
- ArangoDB graph Q&A or writes -> ArangoChatAgent
- CSV-to-knowledge-graph bootstrapping -> CSVGraphAgent

## Route elsewhere

- Unstructured document RAG -> sibling skill `retrieval-doc-chat`
- Provider, model, or API-key setup -> sibling skill `llm-provider-config`
- Generic Task / Tool orchestration -> sibling skill `agents-tasks-tools`

## Start here

- [API reference](references/api-reference.md)
- [Workflow patterns](references/data-agent-workflows.md)
- [Database and schema config](references/database-and-schema-config.md)
- [Security and validation](references/security-and-validation.md)
- [Troubleshooting](references/troubleshooting.md)
- [Safety smoke script](scripts/data_agent_safety_smoke.py)

## Operating rules

- Prefer the least-privilege, validation-first path.
- Keep `TableChatAgentConfig.full_eval=False` unless the input is fully trusted.
- Keep `SQLChatAgentConfig.allow_dangerous_operations=False` and extend
  `allowed_statement_types` only when writes are intended.
- Keep Neo4j and Arango dangerous operations disabled by default.
- Use schema or context helpers before asking the model to write SQL or graph
  queries.
- Use the bundled smoke script to inspect defaults or validate a SQL, Cypher,
  or AQL string locally without touching a live service.
- If the question is really about model/provider configuration, hand it off to
  the provider-config skill instead of forcing it through this one.

Files in this skill

  • SKILL.md2 KB
  • references/api-reference.md3.7 KB
  • references/data-agent-workflows.md3.7 KB
  • references/database-and-schema-config.md3.8 KB
  • references/security-and-validation.md3.6 KB
  • references/troubleshooting.md2.7 KB
  • scripts/data_agent_safety_smoke.py19.7 KB

Attribution

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

Comments

Loading comments…

Newsletter

Get the weekly list of safe new skills, and the ones we pulled.