Agent Portability Checker

Audit and auto-fix agent skills to remove platform lock-in and ensure cross-platform compatibility.

Install
cmdop skills install agensi-agent-portability-checker

Agent Portability Checker is a developer skill that audits agent skill codebases for portability issues that cause platform lock-in. It scans Python-based agent skills for hardcoded paths, platform-specific CLI dependencies, missing environment variable support, and rigid environmental assumptions that prevent skills from running across different agent environments such as Claude Code, OpenClaw, or Codex.

The skill operates against an 8-point checklist that includes XDG Base Directory Specification compliance, cross-platform path resolution using os.path or pathlib, multi-channel formatting compatibility (Slack, Discord, WhatsApp), and headless OAuth compliance. These are checks that are easy to miss in manual code review.

The output is a structured audit report that identifies specific line numbers where issues occur — flagging hardcoded paths, absent environment variable fallbacks, and platform-tied CLI calls. When run in fix mode, the skill automatically refactors problematic paths to use dynamic environment variables such as $SKILL_DATA_DIR instead of leaving static references in place.

This skill is aimed at developers who distribute or plan to distribute agent skills across multiple platforms and want to catch portability problems before deployment rather than debug them after the fact. It is not a general-purpose code linter and does not cover languages other than Python.

Use cases

  • Audit a Python agent skill codebase for hardcoded file paths before publishing to a marketplace
  • Identify missing environment variable support that would prevent a skill from running on a new platform
  • Check XDG Base Directory Specification compliance in an agent skill
  • Auto-refactor hardcoded paths to use dynamic environment variables like $SKILL_DATA_DIR
  • Verify headless OAuth compliance before deploying a skill to a headless agent environment
  • Detect platform-specific CLI dependencies that would break a skill on an unsupported host

When to use it

  • When preparing a Python-based agent skill for deployment across multiple agent platforms
  • When reviewing a skill codebase for technical debt caused by platform-specific assumptions
  • When a skill that works on one platform fails silently on another and the cause is unclear
  • When enforcing portability standards across a team building multiple agent skills

When not to use it

  • When the skill is written in a language other than Python, as only Python-based agent skills are covered
  • When looking for a general-purpose static analysis or security linter
  • When the agent skill is intentionally designed for a single platform and portability is not a goal