Convert Codex Plugin to Claude Code
One-shot conversion of a Codex plugin to Claude Code format. Use when asked to convert a Codex plugin to Claude Code, add CC support for a Codex plugin, or reverse-port a plugin from Codex.
What this skill does
# Convert Codex Plugin → Claude Code
Convert a single plugin from Codex format to Claude Code-compatible format.
## When to activate
- User says "convert X from Codex to CC", "add Claude Code support for X", "reverse-port X"
- User provides a Codex plugin path and asks for CC compatibility
- Round-trip recovery: CC plugin was converted to Codex, now needs CC files restored
## Inputs
- **Plugin path** — relative to repo root, e.g. `plugins/my-codex-plugin`
- **Repo root** — directory containing `.agents/plugins/marketplace.json` (default: detect from cwd)
## Workflow
## Phase 0: Vendor external plugins
When the plugin lives outside this repository, copy it into `plugins_dir`
before conversion. Review license and attribution, resolve name collisions,
discard stale foreign generated artifacts after review, and explicitly choose
the source of truth when attaching. Deterministic scripts never copy files
between repositories.
### Step 1 — Read source
Read the following files from the target plugin (skip if absent):
1. `<plugin-path>/.codex-plugin/plugin.json` — required
2. `<plugin-path>/skills/` directory listing — list all SKILL.md files
3. `<plugin-path>/.mcp.json` — optional
4. `<plugin-path>/README.md` — optional
5. `<plugin-path>/.plugin-cross-port.yaml` — optional (check source_of_truth)
If `.codex-plugin/plugin.json` is missing, stop and report: "Not a Codex plugin: no .codex-plugin/plugin.json found."
If `.plugin-cross-port.yaml` exists with `source_of_truth: claude-code`, warn:
> "This plugin was originally created in Claude Code. Reverse conversion may produce redundant files. Proceed?"
### Step 2 — Check for existing CC manifest
If `<plugin-path>/.claude-plugin/plugin.json` already exists, read it. Ask:
> "`.claude-plugin/plugin.json` already exists. Overwrite or merge?"
### Step 3 — Generate `.claude-plugin/plugin.json`
Build the CC manifest from the Codex manifest:
```json
{
"name": "<same as Codex name>",
"description": "<same description>",
"version": "<same version>",
"author": {
"name": "<author.name>",
"email": "",
"url": ""
},
"keywords": ["<interface.category>", "codex", "cross-platform"],
"license": "MIT",
"skills": [
"./skills/<name>/SKILL.md"
// list all SKILL.md files from skills/
]
}
```
Notes:
- Do NOT generate `hooks` — must be added manually based on plugin logic
- List existing skills explicitly in `skills` array
- `commands` array is only added if Step 5 produces generated commands
### Step 4 — Verify `skills/`
`skills/` is shared between Codex and CC. List all existing skills and confirm they load correctly in CC (SKILL.md format with frontmatter: `name`, `description`).
If a skill has no frontmatter or missing `description`, emit warning.
### Step 5 — Convert `skills/` to CC commands
For each `skills/<name>/SKILL.md` (exclude `generated-from-commands/`):
1. Read SKILL.md frontmatter: `name`, `description`
2. Map `interface.capabilities` → `allowed-tools`:
- `Read` → `Read, Glob, Grep`
- `Write` → `Write, Edit`
- `Execute` → `Bash`
- `Network` → `WebFetch, WebSearch`
3. Write `<plugin-path>/commands/generated-from-codex-<skill-name>.md`:
```markdown
---
description: <description from SKILL.md, without "Use when the user invokes /X." suffix>
argument-hint: "[args]"
allowed-tools: [<mapped tools>]
---
> Converted from Codex skill `<skill-name>`.
> Review and adapt: add specific `allowed-tools` entries and `${CLAUDE_PLUGIN_ROOT}` paths as needed.
<SKILL.md body>
```
If skill body starts with `> Converted from Claude Code command` (round-trip case), strip that header.
Skip generated commands if newer than source skill (idempotent).
### Step 6 — Hooks warning
CC supports hooks (`SessionStart`, `PostToolUse`, `PreCompact`, `Stop`, etc.). Codex has no equivalent.
Emit notice:
> "Hooks are not auto-generated. If this plugin needs event-driven behavior (session tracking, auto-commit, etc.), add hooks manually to `.claude-plugin/plugin.json`."
If `--strict` and `.plugin-cross-port.yaml` doesn't have `decisions.hooks_converted: true`, fail with this message.
### Step 7 — Create/update `.plugin-cross-port.yaml`
```yaml
version: 1
plugin: <plugin-name>
generated_at: <ISO date>
source_of_truth: codex
decisions:
skills_shared: true
skills_converted_to_commands: true
hooks_converted: false # must be set manually after adding hooks
warnings: []
manually_maintained: []
```
### Step 8 — Summary
For standalone one-shot conversion, use:
```bash
python3 plugins/plugin-cross-port/scripts/cross_port.py \
plugin convert <plugin-path> --from codex --to claude-code
```
For a managed repository, attach the plugin after conversion:
```bash
python3 plugins/plugin-cross-port/scripts/cross_port.py \
plugin attach <plugin-path> --source codex
```
```
Plugin Cross-Port (Codex → CC): <plugin-name>
=============================================
Generated:
✅ .claude-plugin/plugin.json
✅ commands/generated-from-codex-<N> commands
✅ .plugin-cross-port.yaml
Shared (no action):
skills/ — <N skills>
Notices:
ℹ️ hooks — not generated, add manually if needed
Manual steps:
1. Review commands/generated-from-codex-*.md — refine allowed-tools.
2. Add hooks to .claude-plugin/plugin.json if needed.
3. Rename/move generated commands to commands/<name>.md when satisfied.
4. Run: python3 scripts/convert_codex_to_cc.py <plugin-path> --repo-root .
to validate idempotency.
```
## Error handling
- Missing `.codex-plugin/plugin.json` → stop with clear error
- `skills/` is empty → warn (or strict fail)
- Skill has no frontmatter → use filename as description, emit warning
- `interface.capabilities` missing → default to `["Read", "Write"]`
Related in Ads & Marketing
ads
IncludedMulti-platform paid advertising audit and optimization skill. Analyzes Google, Meta, YouTube, LinkedIn, TikTok, Microsoft, and Apple Ads. 250+ checks with scoring, parallel agents, industry templates, and AI creative generation.
banana
IncludedAI image generation Creative Director powered by Google Gemini Nano Banana models. Use this skill for ANY request involving image creation, editing, visual asset production, or creative direction. Triggers on: generate an image, create a photo, edit this picture, design a logo, make a banner, visual for my anything, and all /banana commands. Handles text-to-image, image editing, multi-turn creative sessions, batch workflows, and brand presets.
rpg-migration-analyzer
IncludedAnalyzes legacy RPG (Report Program Generator) programs from AS/400 and IBM i systems for migration to modern Java applications. Extracts business logic from RPG III/IV/ILE source code, identifies data structures (D-specs), file operations (F-specs), program dependencies (CALLB/CALLP), and converts RPG constructs to Java equivalents. Generates migration reports, complexity estimates, and Java implementation strategies with POJO classes, JPA entities, and service methods. Use when modernizing AS/400 or IBM i legacy systems, analyzing RPG source files (.rpg, .rpgle, .RPGLE), converting RPG to Java, mapping data specifications to Java classes, planning legacy system migration, or when user mentions RPG analysis, Report Program Generator, RPG III/IV/ILE, AS/400 modernization, IBM i migration, packed decimal conversion, or mainframe application rewrite.
brand-library-architect
IncludedBuild a complete brand library for a product — visual asset render pipeline, brand documentation set (BRAND, COPY, MANIFESTO, BIOS, FAQ, GLOSSARY, TONE, PRICING), open-source convention files (README, CONTRIBUTING, SECURITY, CODE_OF_CONDUCT), and a self-contained press kit. This skill should be used when the user asks to "build a brand library / brand kit / press kit / brand assets" for a product, "set up a brand library workflow," "create a positioning manifesto plus visual identity," or any combination of brand documentation + visual asset pipeline. Apply phase-by-phase or run end-to-end. Templates are product-agnostic and use {{TOKEN}} placeholders the skill prompts the user to fill.
writing-tech-post
IncludedAuthors engineering blog posts end-to-end: launch deep-dives, incident postmortems, architecture migrations, performance case studies, tutorials, AI/agent system writeups, security disclosures, and research-to-product translations. Picks the correct archetype, plans the abstraction ladder, enforces an evidence cadence (diagrams, benchmarks, profiles, traces, code, ablations), tunes voice against publisher house styles (Datadog, Vercel, GitHub, AWS, Meta, Cloudflare, Jane Street), and runs a pre-publish gate for narrative momentum and disclosure ethics. Use when drafting a new engineering post, restructuring a draft that feels flat, deciding which evidence form belongs where, validating that depth and product context are balanced, or preparing a postmortem, migration, or performance narrative for external publication. Do not use for API reference documentation, README authoring, marketing copy, release notes, generic SEO content, ghost-written executive thought leadership, or non-engineering long-form essays.
blog-google
IncludedGoogle API integration for blog performance: PageSpeed Insights, CrUX Core Web Vitals with 25-week history, Search Console performance, URL Inspection, Indexing API, GA4 organic traffic, NLP entity analysis for E-E-A-T, YouTube video search for embedding, and Google Ads Keyword Planner. Progressive feature availability based on credential tier (API key, OAuth/service account, GA4, Ads). Shares config with claude-seo at ~/.config/claude-seo/google-api.json. Use when user says "google data", "page speed", "core web vitals", "search console", "indexation", "GA4", "keyword research", "nlp entities", "blog performance", "youtube search", "google api setup".