frappe-agent-architect
Use when designing multi-app Frappe architectures, deciding whether to split functionality into separate apps, or implementing cross-app communication…
Use when building database queries with frappe.qb in Frappe v14-v16. Covers the PyPika-based query builder: SELECT, INSERT, UPDATE, DELETE, joins, aggregation, subqueries, cross-DB compatibility (MariaDB/PostgreSQL), and migration from raw SQL. Prevents SQL injection and
$ npx -y skills add Impertio-Studio/Frappe_Claude_Skill_Package --skill frappe-syntax-query-builder --agent claude-codeHow it fires
How this skill gets triggered: by you, by Claude, or both.
/frappe-syntax-query-builderContext preview
The summary Claude sees to decide when to auto-load this skill.
Use when building database queries with frappe.qb in Frappe v14-v16. Covers the PyPika-based query builder: SELECT, INSERT, UPDATE, DELETE, joins, aggregation, subqueries, cross-DB compatibility (MariaDB/PostgreSQL), and migration from raw SQL. Prevents SQL injection and
name: frappe-syntax-query-builder description: > Use when building database queries with frappe.qb in Frappe v14-v16. Covers the PyPika-based query builder: SELECT, INSERT, UPDATE, DELETE, joins, aggregation, subqueries, cross-DB compatibility (MariaDB/PostgreSQL), and migration from raw SQL. Prevents SQL injection and DB-specific bugs. Keywords: frappe.qb, query builder, DocType, Field, pypika, join,, frappe.qb example, how to write query, join tables, SQL replacement, parameterized query. aggregate, ImportMapper, cross-database, parameterized query. license: MIT compatibility: "Claude Code, Claude.ai Projects, Claude API. Frappe v14-v16." metadata: author: OpenAEC-Foundation version: "3.0"
from frappe.query_builder import DocType, Field from frappe.query_builder.functions import Count, Sum, IfNull from frappe.query_builder.custom import ConstantColumn, GROUP_CONCAT from frappe.query_builder.terms import SubQuery from frappe.query_builder.utils import ImportMapper, db_type_is from pypika.terms import Case, ValueWrapper from pypika import CustomFunction, Order
| Action | Pattern | |--------|---------| | SELECT | `frappe.qb.from_("DocType").select("field1", "field2")` | | WHERE | `.where(Field("status") == "Open")` | | ORDER | `.orderby("creation", order=Order.desc)` | | LIMIT | `.limit(10).offset(0)` | | JOIN | `.left_join(dt2).on(dt2.name == dt1.parent)` | | COUNT | `.select(Count("*"))` | | INSERT | `frappe.qb.into("DocType").columns("f1", "f2").insert("v1", "v2")` | | UPDATE | `frappe.qb.update("DocType").set("field", "value").where(...)` | | DELETE | `frappe.qb.from_("DocType").delete().where(...)` | | RUN | `.run()` (tuples) / `.run(as_dict=True)` (dicts) |
---
Need to query the database?
│
├─ Simple get/list → frappe.db.get_value(), frappe.get_all()
│ (See frappe-core-database skill)
│
├─ Complex query with joins/aggregates/subqueries → frappe.qb ✓
│
├─ Very complex SQL not expressible in qb → frappe.db.sql()
│ (ALWAYS use parameterized values: frappe.db.sql(query, values))
│
└─ Need cross-DB compatibility → frappe.qb + ImportMapper ✓
Using frappe.qb?
│
├─ Table reference → DocType("Sales Order") — NEVER use Table()
├─ Field reference → dt.field_name or Field("field_name")
├─ Execute → .run() for tuples, .run(as_dict=True) for dicts
├─ DB-specific function → ImportMapper({db_type_is.MARIADB: X, db_type_is.POSTGRES: Y})
└─ Get SQL string → query.get_sql() — NEVER pass to frappe.db.sql()---
# ALWAYS use DocType() for table references — adds "tab" prefix
so = frappe.qb.DocType("Sales Order")
soi = frappe.qb.DocType("Sales Order Item")
orders = (
frappe.qb.from_(so)
.select(so.name, so.customer, so.grand_total)
.where(so.status == "To Deliver and Bill")
.where(so.docstatus == 1)
.orderby(so.creation, order=Order.desc)
.limit(20)
.run(as_dict=True)
)result = (
frappe.qb.from_(so)
.left_join(soi).on(soi.parent == so.name)
.select(so.name, so.customer, soi.item_code, soi.qty)
.where(so.docstatus == 1)
.where(soi.item_code.like("ITEM-%"))
.run(as_dict=True)
)from frappe.query_builder.functions import Count, Sum
gl = frappe.qb.DocType("GL Entry")
result = (
frappe.qb.from_(gl)
.select(gl.account, Sum(gl.debit).as_("total_debit"), Count("*").as_("entries"))
.where(gl.docstatus == 1)
.groupby(gl.account)
.run(as_dict=True)
)
# Shortcut aggregation methods
total = frappe.qb.sum("GL Entry", "debit", filters={"account": "Sales"})
max_qty = frappe.qb.max("Stock Ledger Entry", "actual_qty", filters={"item_code": "ITEM-001"})# INSERT
frappe.qb.into("Activity Log").columns("user", "action").insert("admin", "login").run()
# UPDATE
customer = frappe.qb.DocType("Customer")
(frappe.qb.update(customer)
.set(customer.status, "Active")
.where(customer.name == "CUST-001")
.run())
# DELETE
frappe.qb.from_("Error Log").delete().where(Field("creation") < "2024-01-01").run()---
dt = frappe.qb.DocType("Sales Order")
# Equality
.where(dt.status == "Open")
# OR (pipe operator)
.where((dt.status == "Open") | (dt.status == "Draft"))
# AND (chain .where() calls)
.where(dt.status == "Open")
.where(dt.docstatus == 1)
# LIKE
.where(dt.customer.like("CUST-%"))
# IN
.where(dt.status.isin(["Open", "Draft"]))
# BETWEEN (bracket syntax)
.where(dt.creation[start_date:end_date])
# NULL checks
.where(dt.email.isnotnull())
.where(dt.phone.isnull())
# Comparison
.where(dt.grand_total > 1000)
.where(dt.grand_total >= 500)
.where(dt.status != "Cancelled")---
from frappe.query_builder.utils import ImportMapper, db_type_is
from frappe.query_builder.custom import GROUP_CONCAT, STRING_AGG
# ImportMapper selects correct function per database
GroupConcat = ImportMapper({
db_type_is.MARIADB: GROUP_CONCAT,
db_type_is.POSTGRES: STRING_AGG,
})
dt = frappe.qb.DocType("Has Role")
result = (
frappe.qb.from_(dt)
.select(dt.parent, GroupConcat(dt.role))
.groupby(dt.parent)
.run(as_dict=True)
)| MariaDB | PostgreSQL | Use ImportMapper | |---------|-----------|-----------------| | `GROUP_CONCAT` | `STRING_AGG` | Yes | | `MATCH...AGAINST` | `TO_TSVECTOR` | Yes | | `Locate` | `Strpos` | Yes | | `Timestamp` | Extract-based | Auto-handled |
---
1. **NEVER pass qb query to `frappe.db.sql()`** — bypasses parameterization 2. **NEVER use `Table()` for DocTypes** — use `DocType()` (adds `tab` prefix) 3. **NEVER forget `.run()`** — without it you get a query object, not results 4. **NEVER use raw SQL strings in `frappe.get_all(fields=[...])`** — use dict syntax 5. **ALWAYS use `ImportMapper` for DB-specific functions** 6. **ALWAYS chain `.run(as_dict=True)
60 deterministic Claude AI skills for Frappe Framework & ERPNext v14-v16 development and operations
Repo: Impertio-Studio/Frappe_Claude_Skill_Package
Use when designing multi-app Frappe architectures, deciding whether to split functionality into separate apps, or implementing cross-app communication…
Use when debugging Frappe errors, using bench console for live inspection, analyzing tracebacks, or reading Frappe log files. Prevents wasted debugging time…
Use when receiving vague or unclear ERPNext/Frappe development requests that need interpretation. Transforms requirements like 'make invoice auto-calculate' or…
Use when migrating a Frappe app between major versions, detecting breaking API changes, or resolving post-migration errors. Prevents failed migrations from…
Use when reviewing or validating Frappe/ERPNext code against best practices and common pitfalls. Checks generated code before deployment, validates against all…
Use when building ERPNext/Frappe API integrations (v14/v15/v16) including REST API, RPC API, authentication, webhooks, and rate limiting. Covers external API…