agentleFS
KhaledAufDBKhaledAufDB/lakeforge/.claude/skills/forge-review/SKILL.md

forge-review

Skill0 starsChanged 2 months agoMITView on GitHub

Review a built source integration against its plan and the expert rules. Produces a PASS/FAIL report with findings by severity.

12 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use forge-ship and Forge.

Other files people use alongside this one.

Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
  • redteamlifeSkill

    forge-ship

    Use when the user asks to merge, release, promote, reconcile, or close out work in a repo containing `docs/forge/` or governed by FORGE. Not for per-task commits; those complete inside forge-execute-task.

    Also in 5 projects that use this

  • ikennaokpalaSkill

    Forge

    Autonomous quality engineering swarm that forges production-ready code through continuous behavioral verification, exhaustive E2E testing, and self-healing fix loops. Combines DDD+ADR+TDD methodology with BDD/Gherkin specifications, 7 quality gates, defect prediction, chaos testing, and cross-context dependency awareness. Architecture-agnostic - works with monoliths, microservices, modular monoliths, and any bounded-context topology.

    Also in 8 projects that use this

The file

We split this skill into the rules it gives an agent

2 rules. 0 are common, 0 are distinctive, and 2 are very specific to this project.

  • 0

    Common

    Most of the 9 edited versions we read keep these.

  • 0

    Distinctive

    This author's own choices. Worth a look.

  • 2

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Read every file under sources/<source-name>/.

    line 18
  2. Very specific:

    Read .claude/commands/experts/databricks.md sections on MERGE, watermarks, timestamps, numerics, clustering — the review is against them, …

    line 19

“Common” is measured: kept by at least 6 in 10 of the 9 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.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Forge Review

You are a senior data architect reviewing an integration for correctness, completeness and conformance. Findings are specific (file, line, what, why) — not a restatement of the checklist.

Input

/forge-review <source-name>

Steps

  1. Read specs/<source-name>-integration-plan.md.
  2. Read every file under sources/<source-name>/.
  3. Read .claude/commands/experts/databricks.md sections on MERGE, watermarks, timestamps, numerics, clustering — the review is against them, not against memory.
  4. Validate against the checklist below. Grep for leftovers from the example source (erp, sales_order) and TODO.

Review checklist

Plan fidelity

  • Every planned bronze, silver and gold object has its config entry and notebook
  • Column names, types and grains match the plan; deviations are listed in the plan's §12 or flagged
  • Every gold column has a reason in the plan; no audit/operational columns leaked into gold

Conformance

  • No new dimension duplicates a registry entry
  • FK resolution mode matches the plan (hash only where the id namespace is shared)
  • source_system_id is the registered one and appears on every gold row

MERGE and keys

  • Merge keys equal the plan's declared grain; the source business key is among them
  • WHEN MATCHED is gated with IS DISTINCT FROM (or build_merge_sql is used)
  • Nullable merge keys use <=>
  • Multi-version CDF input is deduplicated to one row per key before the MERGE
  • No F.lit(None) without a cast; no try: merge except: overwrite

Types and time

  • Timestamps are UTC *_at; no TIMESTAMP WITH TIME ZONE; DATE where a date suffices
  • Decimals are sized (DECIMAL(p,s)); no bare DECIMAL; floats sized to the magnitude they can reach
  • Ratios stored 0–1; durations carry a unit suffix

Incremental correctness

  • Watermark column is a bare indexed source column; lookback is set for mutable sources
  • A second run over unchanged input writes 0 rows (the idempotency test is in the test plan)

Layout and performance

  • cluster_by ≤ 4 columns, most-filtered first, time-first for event streams
  • No correlated scalar subquery in a projection; no CTE referenced more than once on the hot path

Security

  • No credential values anywhere; only secret-scope key names
  • PII columns identified in the plan and handled (mask, tag, or explicitly cleared)

Deployment

  • deploy.py --dry-run produces the planned task graph
  • Tests cover every gold table: row count, unique grain, RI, FK resolution rate

Output

# <Source> Integration Review
## Verdict: PASS | FAIL
## Findings
### Critical (must fix)     — file:line — what — why (link the expert section)
### Warnings (should fix)
### Info
## Files reviewed
## Recommendation

Download now (ZIP)

Does 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
What did you use it for? e.g. Reviewing pull requests in a Rails app

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.