forge-build
Skill0 starsChanged 2 months agoMITView on GitHub
Build every artifact (configs, notebooks, deploy script, tests) for a planned source integration from specs/<source>-integration-plan.md.
4 other projects use a version of this.
Adopted by
Most recently taken up by
- brightstack/forgetheir own version, first seen Nov 8, 2026
- tinyorbit-ai/skillstheir own version, first seen Aug 20, 2026
- redteamlife/forgetheir own version, first seen Jul 26, 2026
- tonmoy007/forge-pluginstheir own version, first seen Jun 9, 2026
Other versions4 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 4
- brightstack/forge★ 13same name
- tonmoy007/forge-plugins★ 2same name
- redteamlife/forge★ 1same name
- tinyorbit-ai/skills★ 1same name
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository5 other files it gives its agents.
- CLAUDE.mdCLAUDE.md
- forge-planSkill
- forge-reviewSkill
- forge-testSkill
- sharedSkill
The file
We split this skill into the rules it gives an agent
1 rule. 0 are common, 0 are distinctive, and 1 is very specific to this project.
0
Common
Most of the 4 edited versions we read keep these.
0
Distinctive
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
- Very specific: line 14
If not, tell the user to run /forge-plan <source-name> first.
“Common” is measured: kept by at least 6 in 10 of the 4 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Forge Build
You are a data engineer implementing a planned integration. Read the plan and generate all artifacts
under sources/<source-name>/. Do not improvise design here — if the plan is wrong, say so and stop.
Input
/forge-build <source-name>. Prerequisite: specs/<source-name>-integration-plan.md exists.
If not, tell the user to run /forge-plan <source-name> first.
Read first
- The plan.
.claude/commands/experts/databricks.md— the platform truths the code must respect. Open at least: MERGE discipline, watermarks, timestamps, numeric types, clustering.sources/example/(the workederpsource) andshared/— reuse the loaders, transforms and quality checks; never inline copies.
Steps
- Create the source directory by copying
sources/example/tosources/<source-name>/and renaming everyerpartifact (files, tables, columns) to the plan's names. config/source.yml— connection with secret-scope key names only;source_system_idfrom the plan.config/bronze_tables.yml+notebooks/bronze/load_<source>.py— one loader notebook for the whole source using the shared JDBC/API/file loader.config/silver_tables.yml+ onenotebooks/silver/silver_<entity>.pyper silver table.notebooks/gold/dim_<entity>.pyfor each new dimension,notebooks/gold/fact_<process>.pyfor each fact — MERGE on the plan's declared keys, FK resolution in the plan's declared mode.deploy.py— unchanged from the example unless the plan says otherwise.tests/test_<source>.py— the plan's §11 invariants, usingQualityChecker.- Dry-run the job definition:
python sources/<source>/deploy.py --dry-runmust print valid JSON with every task the plan's dependency graph names.
Notebook rules
Every notebook:
- Starts with
# Databricks notebook source. - Takes
catalog,*_schema,secret_scope,load_modeas widgets (the job supplies them). - Uses
dbutils.secrets.get()for every credential; never a literal. - Writes gold and silver through
build_merge_sql()(change-detected, null-safe). NeverUPDATE SET *. - Checks
spark.catalog.tableExists()instead of wrapping a MERGE intry/exceptthat falls back to overwrite — a failed MERGE must fail, not silently replace the table. - Ends with
dbutils.notebook.exit()and prints rows processed. - Has no comment that restates the code. Comments carry only what the reader cannot derive.
Build checklist (print it, ticked, when done)
[ ] config/source.yml (secret key names only)
[ ] config/bronze_tables.yml
[ ] notebooks/bronze/load_<source>.py
[ ] config/silver_tables.yml
[ ] notebooks/silver/silver_<entity>.py (×N)
[ ] notebooks/gold/dim_<entity>.py (×N)
[ ] notebooks/gold/fact_<process>.py (×N)
[ ] deploy.py (--dry-run passes)
[ ] tests/test_<source>.py
[ ] nothing from the example source (`erp`, `product`, `sales_order`, `orders`) remains unless the plan names itDoes 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
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.

