skillsFirst-partyReview first Safety · Privacy ·
Codex Plugin Creator Capability Pack Skill
Expert Codex plugin capability pack for safe plugin scaffolding, manifest quality, MCP integration, and maintainable distribution.
by JSONbored·added 2026-04-11·83,749 source repo stars·
Claude CodeCodexWindsurfGeminiCursorCLI
HarnessClaude CodeCodexWindsurfGeminiCursorCLI
Level:expertType:capability-packVerified:validated
Review first — review before installing
Open the source and read safety notes before installing.
Prerequisites
- Plugin use case and scope
- Manifest and MCP requirements
- Deployment target constraints
Schema details
- Install type
- package
- Reading time
- 9 min
- Difficulty score
- 87
- Troubleshooting
- Yes
- Breaking changes
- No
Source repository stats
- Scope
- Source repo
- Stars
- 83,749 source repo stars
- Forks
- 12,151
- Updated
- 2026-05-19T11:39:05Z
Package metadata
- Package verified
- Yes
- SHA-256
- 7ea59a73540f97320dc9819720d4dc25f6d117f44096d7aa6011a4857cced8f0
Skill and platform metadata
- Skill type
- capability-pack
- Skill level
- expert
- Verification
- validated
- Verified at
- 2026-04-11
Retrieval sources
https://github.com/openai/codexhttps://github.com/modelcontextprotocolhttps://spec.modelcontextprotocol.io/specification/
Tested platforms
ClaudeCodexOpenClawCursorWindsurfGemini
| Platform | Support | Install path |
|---|---|---|
| claude-code | Native | .claude/skills/<skill-name>/SKILL.md |
| codex | Native | .agents/skills/<skill-name>/SKILL.md |
| windsurf | Native | .windsurf/skills/<skill-name>/SKILL.md |
| gemini | Native | .gemini/skills/<skill-name>/SKILL.md or .agents/skills/<skill-name>/SKILL.md |
| cursor | Adapter | .cursor/rules/<skill-name>.mdc |
| cli | Manual | AGENTS.md or tool-specific context file |
Full copyable content
# Trigger
"Apply the Codex plugin creator capability pack."
# Required output
1) Plugin architecture and capability boundaries
2) Manifest and permissions review
3) Security and observability checklist
4) Release and maintenance planAbout this resource
Knowledge Freshness
This capability pack is pinned to documentation verified on 2026-04-11.
Retrieval Sources
- https://github.com/openai/codex
- https://github.com/modelcontextprotocol
- https://spec.modelcontextprotocol.io/specification/
Core Workflow
- Define plugin scope, trust boundaries, and runtime dependencies.
- Build manifest with minimum required capabilities only.
- Validate tool behavior and failure paths with regression tests.
- Add observability and troubleshooting hooks.
- Publish with maintenance and compatibility policy.
Overview
This skill helps agents build practical plugins that stay secure and maintainable instead of becoming one-off brittle integrations.
Capability Scope
- Plugin architecture and capability boundaries
- Manifest quality and permission minimization
- MCP tool interoperability guidance
- Error handling and observability setup
- Release governance and compatibility notes
Compatibility
Native
- Claude Code / Claude: native skill usage via
SKILL.md. - Codex/OpenAI workflows: compatible with Agent Skills-style
SKILL.mdcontent as reusable workflow instructions.
Manual Adaptation
- Gemini CLI: native skill usage via
.gemini/skills/<skill-name>/SKILL.mdor.agents/skills/<skill-name>/SKILL.mdwhere supported. - Cursor: use the generated
.cursor/rules/*.mdcadapter for project rules. - OpenClaw and similar agents: use the same skill content as a reusable prompt/workflow file when native skill import is unavailable.
Production Rules
- Minimize plugin permissions and document each permission reason.
- Validate tool contracts against strict schema expectations.
- Keep plugin error handling explicit and observable.
- Ship updates with backward-compatibility notes when behavior changes.
Troubleshooting
Issue: Plugin has excessive privilege scope
Fix: reduce manifest permissions to explicit operational minimum.
Issue: Tool calls are inconsistent across clients
Fix: enforce contract tests against expected schema/shape.
Issue: Plugin errors are hard to diagnose
Fix: add request IDs, structured logs, and clear failure categories.
Output Contract
- Plugin plan with trust boundaries.
- Manifest with justification per permission.
- Validation and regression-test protocol.
- Maintenance and deprecation plan.
Content outline
#codex#plugins#mcp#developer-tooling#capability-pack
Source citations
Signals
Loading live community signals…
More like this, weekly
A short, calm digest of reviewed Claude resources. Unsubscribe any time.