Generate pg_durable SQL graphs for PostgreSQL workflows with checkpointing.
npx skills add https://github.com/microsoft/pg_durable --skill pg-durable-sql
Please help me install this Agent Skill. Skill: pg-durable-sql Source: https://github.com/microsoft/pg_durable/tree/main/.agents/skills/pg-durable-sql Command: npx skills add https://github.com/microsoft/pg_durable --skill pg-durable-sql
Durable SQL graph generation and orchestration for PostgreSQL workflows, enabling checkpointing and seamless resume after crashes without external services.
Create a simple durable workflow by composing SQL nodes and starting it with df.start.
High-intent search queries and answers about installing and using this skill.
Durable PostgreSQL workflows use a SQL-based DSL with df.* operators to build execution graphs, enabling checkpointing and seamless resume after crashes without external services. You compose nodes using df.start to initiate the workflow.
Variable substitution in SQL workflow orchestration requires setting variables via df.setvar before starting the workflow with df.start. Result captures use |=> syntax, and system variables like {sys_*} are supported for dynamic value passing across nodes.
PostgreSQL SQL workflows support parallel fan-out/fan-in, race conditions, conditional branches, and loops within the DSL graph. These constructs enable complex ETL pipelines, event-driven micro-workflows, and scheduled maintenance tasks directly inside the database.
DSL expressions in durable SQL graphs must be TEXT type, with variables set before df.start and proper escaping applied. The Skill enforces these rules and provides a complete operator and function reference for building production-grade workflows.
No external orchestration services are needed for scheduled maintenance and ETL pipelines in PostgreSQL. The durable SQL DSL supports cron scheduling, HTTP calls, and signals natively, enabling checkpointing and crash recovery entirely within the database.
Durable SQL workflows require all variables to be initialized via df.setvar before calling df.start. This ensures proper state checkpointing and variable substitution during execution, preventing failures caused by undefined references in the workflow graph.
JeffallanOptimize SQL queries and schemas across PostgreSQL, MySQL, SQL Server, and Oracle.
masharrattAutomate CFN Loop sprint planning, execution, and checkpoint recovery.
bobmatnycManage multiple Git worktrees for parallel development across branches.
CatsofsufferingManage and validate data pipelines and streaming workflows across orchestration and quality tools.
