Review a plugin, skill bundle, or agent workflow that has been moved from a standalone repository into a marketplace layout. Use this when you need to verify path rewrites, manifest identity, registry metadata, changelog/version drift, hook runtime assumptions, copied assets, tests, examples, utilities, and migration notes before publishing a marketplace package. Ask for a "marketplace migration review", "plugin import audit", "path adaptation check", or "marketplace readiness review" to use ...
Scanned 9/6/2026
Install to Claude Code
npx -y skills add fakoli/fakoli-plugins --skill marketplace-migration-review --agent claude-codeInstalls into .claude/skills of the current project.
Are you the author of Marketplace Migration Review?
Add the live security badge to your README — it updates automatically with every re-scan.
[](https://www.skillsdirectory.com/skills/fakoli-marketplace-migration-review)More formats (shields.io, HTML) on the badges page.
---
name: marketplace-migration-review
description: >
Review a plugin, skill bundle, or agent workflow that has been moved from a standalone
repository into a marketplace layout. Use this when you need to verify path rewrites,
manifest identity, registry metadata, changelog/version drift, hook runtime assumptions,
copied assets, tests, examples, utilities, and migration notes before publishing a
marketplace package. Ask for a "marketplace migration review", "plugin import audit",
"path adaptation check", or "marketplace readiness review" to use this workflow.
---
In Codex, first read the [native runtime adapter](../../references/codex-runtime.md).
# Marketplace Migration Review
## When to Use
Use Marketplace Migration Review when:
- A plugin, agent bundle, or skill set has been copied from a standalone repository into a marketplace repository.
- You need to confirm that manifests, agent definitions, skills, hooks, tests, utilities, examples, and assets still resolve from the new location.
- You need a compact readiness report before publishing or merging a marketplace plugin change.
- A migration has subtle old-to-new path assumptions, such as `.claude/agents/` becoming `agents/` or `reference/` becoming a plugin-local path.
Do **not** use Marketplace Migration Review when:
- You only need to evaluate a system architecture or vendor proposal. Use `complexity-mapper` or `architecture-risk-review` instead.
- You are creating a new product requirements document. This skill reviews migration integrity, not product scope.
- There is no source layout to compare against and no migration assumptions to inspect.
## Inputs Required
| Input | Required | Description |
| --- | --- | --- |
| Marketplace plugin path | Yes | The destination plugin directory, such as `plugins/systems-thinking`. |
| Source layout or provenance | No | A local upstream checkout, archive, or documented source repository. Strongly recommended when checking dropped files or asset parity. |
| Migration record | No | A `MIGRATION.md`, release note, or PR description that claims which paths and assumptions were adapted. |
| Validation commands | No | Repo-specific validation, registry drift, changelog, and test commands that should pass before publishing. |
| Known exclusions | No | Files intentionally not imported, such as standalone CI workflows, setup scripts, or source-repo scaffolding. |
## Process Steps
### Step 1: Build the Migration Inventory
Run `doc-indexer` over the source layout, destination plugin layout, and any migration record. Produce an inventory that separates:
1. Runtime definitions: manifests, agents, skills, hooks, commands, settings.
2. Supporting material: docs, specs, references, examples, fixtures, utilities, assets.
3. Repository scaffolding: CI workflows, setup scripts, lint config, package metadata, registry output.
Record source and destination paths for every runtime definition and any supporting material that is referenced by runtime definitions.
### Step 2: Check Manifest And Registry Identity
Inspect marketplace metadata and compare it with upstream identity assumptions:
- Plugin manifest name, version, repository URL, license, keywords, and description.
- Marketplace aggregate files such as root marketplace metadata and registry indexes.
- Version markers such as `VERSION`, package metadata, lockfiles, and changelog headings.
- Install instructions and README identity references.
Flag any identity split where runtime names, docs, hooks, tests, or registry metadata disagree.
### Step 3: Trace Path Adaptations
Use `doc-reader` and `caveat-extractor` to trace every path or runtime assumption that changed during import:
- Agent definition directories, tool frontmatter, and agent invocation names.
- Skill references to `reference/`, examples, prompts, vendor docs, previous designs, or output contracts.
- Hook command paths, environment variables, payload fields, transcript matching, and plugin invocation prefixes.
- Utility imports, path discovery, CLI help text, and generated output locations.
- Test fixture paths, eval harness paths, assets, README image links, and examples.
Every adapted path should have an explicit old-to-new mapping in the migration record or PR evidence. If the migration record is missing a mapping, list it as a gap.
### Step 4: Verify Behavior And Evidence
Run the available deterministic checks from the marketplace repository root. Prefer the repository's own scripts when available:
```bash
./scripts/validate.sh plugins/<plugin-name>
./scripts/test-path-resolution.sh plugins/<plugin-name>
./scripts/check-changelogs.sh
./scripts/check-registry-drift.sh
(cd plugins/<plugin-name> && uv run pytest tests -q)
```
If the full test suite intentionally excludes evals or slow tests by default, state that clearly and include the collection count or marker behavior.
### Step 5: Synthesize A Readiness Report
Pass the inventory, path trace, validation output, and unresolved gaps to `synthesis-brief-writer`. The final report should distinguish confirmed evidence from inferred readiness.
## Output Format
Produce a **Marketplace Migration Review**:
```markdown
## Marketplace Migration Review
### Executive Summary
[3-5 sentences covering readiness, main risks, and whether publication is recommended]
### Migration Inventory
| Area | Source Location | Marketplace Location | Status |
| --- | --- | --- | --- |
| Manifest | [path] | [path] | [preserved/adapted/missing] |
### Adaptation Map
| Adapted Assumption | Original | Marketplace | Evidence |
| --- | --- | --- | --- |
| [Example: agent directory] | [.claude/agents] | [agents] | [file/line or command] |
### Validation Results
| Command | Result | Notes |
| --- | --- | --- |
| [command] | [pass/fail/not run] | [important output or limitation] |
### Gaps And Risks
- [Gap]: [why it matters, source anchor, recommended action]
### Recommendation
[Ship / ship with caveats / block], with next actions in priority order.
```
## Failure Modes and Caution Points
| Failure Mode | Signal | Response |
| --- | --- | --- |
| Source layout unavailable | No upstream checkout or archive exists for comparison | Review destination consistency and migration record, but label dropped-file confidence as limited. |
| Migration record is too vague | It lists categories but not concrete old-to-new paths | Treat this as a documentation gap and require explicit mappings for runtime definitions. |
| Registry drift | Manifest version differs from registry, lockfile, or changelog | Regenerate registry metadata and align version markers before publishing. |
| Hook runtime mismatch | Hook JSON points to stale paths or old payload field names | Verify against current hook payload schema and run hook-focused tests. |
| Tests pass only from repo root | Tests fail when run from the plugin directory | Fix relative paths or project metadata so the plugin validates from `plugins/<plugin-name>`. |
| Asset parity assumed but not checked | Image or binary files exist but were not checksummed | Compare file lists and checksums before declaring assets preserved. |
## Related Agents
- `doc-indexer` maps the source and marketplace layouts.
- `doc-reader` extracts concrete path and runtime assumptions.
- `caveat-extractor` identifies stale paths, missing evidence, and hidden assumptions.
- `architecture-dependency-mapper` can map runtime dependencies when hooks, utilities, and tests interact.
- `synthesis-brief-writer` turns the evidence into the final readiness report.
Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.
No comments yet. Be the first to comment!