# Agent Prompts

> Ready-to-use prompts for pREST tasks in OpenClaw. Use when the user wants a starting prompt for configs, SQL routes, security review, MCP setup, or project explanation.

- Skill: `prest/agent-prompts` (Agent Skill)
- Install (CLI): `npx skillmds@latest add prest/agent-prompts`
- Raw SKILL.md: https://api.skillmd.com/api/skills/prest/agent-prompts/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Data & Analytics
- Author: prest (https://skillmd.com/u/prest)
- Updated: 2026-09-17
- Page: https://skillmd.com/skills/prest/agent-prompts

---


# pREST agent prompts

Copy-paste prompts for common pREST workflows:

## Generate configuration

> Generate a pREST config for this PostgreSQL database.

Provide schema details, auth needs, and whether they use Docker or bare `prestd`.

## SQL template route

> Create a SQL template route for this query.

Include the SQL, desired HTTP verb, route folder/name, and parameters.

## Security review

> Review this pREST config for security risks.

Point the agent at `prest.toml`, `.env.example`, `queries/`, and Docker Compose if present.

## MCP setup

> Set up pREST MCP with a local pREST instance.

Cover `prest-mcp`, `PREST_MCP_URL=http://localhost:3000/_mcp`, optional `PREST_MCP_TOKEN`, and registry `io.github.prest/prest`.

## Read-only Postgres user

> Create a read-only PostgreSQL user for AI/agent workflows.

Request least-privilege `GRANT SELECT` on needed schemas/tables only.

## Explain project

> Explain this pREST project structure.

Ask the agent to walk through `prest.toml`, query layout, auth, expose/access settings, and MCP wiring if any.