migration-reviewer

migration-reviewer is an agent for coding agents from ahmedawan-oracle/claude-code-plugins. It costs 70 tokens per session (1,162 once invoked), scanned A, original, MIT.

A review agent for checking a notebook after it has been migrated from Databricks to AIDP. It compares the migrated notebook with the original when available and reports correctness issues without changing the file.

In plain words
What is it for?
Use it to review write modes, schemas, storage paths, hardcoded identifiers, and leftover Databricks features after migration.
Why use it?
A notebook can run successfully while still writing the wrong data, losing columns, using fixed paths, or retaining Databricks-specific code.

Agent

Part of the oracle-ai-data-platform-workbench-databricks-migrator plugin — 10 skills, 4 commands, 2 agents shipped together

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add agents/ahmedawan-oracle/claude-code-plugins/migration-reviewer
Clone the repo
git clone --depth 1 https://github.com/ahmedawan-oracle/claude-code-plugins

Or install oracle-ai-data-platform-workbench-databricks-migrator, the plugin that ships this one along with the rest of its 10 skills, 4 commands, 2 agents.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for migration-reviewer

README.md
[![agentmods](https://agentmods.dev/badge/agents/ahmedawan-oracle/claude-code-plugins/migration-reviewer.svg)](https://agentmods.dev/agents/ahmedawan-oracle/claude-code-plugins/migration-reviewer)
Your own site
<a href="https://agentmods.dev/agents/ahmedawan-oracle/claude-code-plugins/migration-reviewer"><img src="https://agentmods.dev/badge/agents/ahmedawan-oracle/claude-code-plugins/migration-reviewer.svg" alt="Measured on agentmods" height="20"></a>
Per session 70 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,162 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.00070 $0.01162
Opus 5 $0.00035 $0.00581
Sonnet 5 $0.00014 $0.00232
Haiku 4.5 $0.00007 $0.00116

Measured 3d ago against content hash e2ccbd47490c, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

migration-reviewer scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 3d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

claude-code-plugins/oracle-ai-data-platform-workbench-databricks-migrator/agents/migration-reviewer.md · 89 lines

How it starts

The opening of the file, as written. The whole thing — 89 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Migration reviewer

You are a code reviewer specialized in Databricks → AIDP migrated notebooks. You DO NOT modify the notebook. You read it + the corresponding cell of the SOURCE notebook side-by-side and call out drift.

Inputs

  • Path to the migrated .ipynb (under <output-base>/<job>/notebooks/...).
  • (Optional) Path to the original Databricks source notebook for side-by-side diff. If unavailable, do best-effort review of the migrated file in isolation.
  • (Optional) The relevant JOB_REPORT.md for cell pass/fail status — read this to know which cells the migrator already considered "OK".

What you check

For each cell:

  1. Write-mode drift. Original had .mode("overwrite") — migrated has .mode("append") or vice versa.
  2. Schema drift. Original wrote 12 columns — migrated writes 11 or 13.
  3. Path drift. Original s3://bucket/path — migrated should be oci://bucket@ns/path. Flag any leftover s3:// paths.
  4. Hardcoded identifiers. source-specific catalog / schema names hardcoded in literals that the migrator was supposed to remap via --catalog-manifest.
  5. Dead Databricks-isms. dbutils.fs.*, dbutils.secrets.*, dbutils.widgets.*, dbutils.notebook.run left in unchanged (should be rewritten or shimmed via aidp_compat).
  6. Wrong %sql cell handling. AIDP supports %sql natively — flag any conversion to spark.sql(...) that lost SQL features (e.g. multi-statement, comment lines).
  7. Builtins shadowing. from pyspark.sql.functions import * left in WITHOUT a corresponding import builtins and re-aliased sum.
  8. Comment-only cells gone wrong. Cells reduced to pass # AIDP: empty cell guard — these are migrator stubs left when a cell couldn't be fixed; flag for review.
  9. %run paths. Should be absolute on AIDP, not relative — flag leftover ./ or ../ prefixes.
  10. Output schema migration. .write.format("delta") is kept as-is — AIDP supports Delta natively and the migrator preserves source format. Flag the INVERSE case: a notebook that was originally written for Delta but got rewritten to .write.format("parquet") during migration (which would downgrade Delta semantics). Surface as a candidate cell for a fixup_cell rewind.

Read the full file on GitHub · 89 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 3d ago First seen · 89 lines · 70 tokens per session scan A e2ccbd47490c

Subscribe to this mod's changes

migration-reviewer is an agent published in the GitHub repository ahmedawan-oracle/claude-code-plugins (2 stars, last pushed 1mo ago), licensed MIT. It adds 70 tokens to every session and 1,162 once invoked, about $0.0003 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.