- Home
- Skills
- Agents & Orchestration
- Sync Agent Skills Across Claude, Codex & Copilot
Sync Agent Skills Across Claude, Codex & Copilot
Keep one skills folder in sync with .codex, .opencode and .github. Read-only drift report, CI check, CLAUDE.md/AGENTS.md symlink.
$7
Works with the AI Agents you already use
Sync Agent Skills Across Claude, Codex & Copilot
Example session with this skill installed
I keep my skills in .claude/skills and copy them to .codex by hand. Sync them, and give me a CI check.
- Read your context and instructions
- Compiled the sync agent across
- Generated the document
Summary
Drift report written to skill-sync-report.md (example: 2 missing, 1 changed, 1 extra in .codex). Nothing is written until you confirm.
What you get
- A drift report per target: in-sync, added, changed, extra (only in the target) or unsupported
- A confirmation question before any file is changed, with a diff for anything edited only in the target
- Updated
.codexcopies of whole skill folders (scripts and reference files included) scripts/sync-skills(Python 3, no network calls) with--checkfor CI and--apply- A list of files to commit
Next steps
- Review
skill-sync-report.mdand confirm the canonical source - Approve the sync, then add
scripts/sync-skills --checkto your CI
sync-agent-skills-across-claude-codex-co.pdf
PDF · document
Example file from a real run - the skill writes it into your workspace.
Connects securely to your tools. The creator never sees your data.
About this skill
What it does
skill-portability-sync keeps one canonical set of skills (default .claude/skills) in sync across the agent folders of a repo: .codex, .opencode and .github. It also keeps CLAUDE.md and AGENTS.md aligned.
- Discovers which agent folders and skills exist, then confirms the canonical source with you.
- Read-only drift check: every skill per target is marked in-sync, added, changed, extra (only in the target) or unsupported. Line endings and trailing whitespace are ignored.
- Writes
skill-sync-report.md, then asks for your confirmation before changing any file. - Copies whole skill folders (scripts and reference files included, file modes preserved). It never overwrites a target-only edit without showing the diff and asking.
- Checks whether CLAUDE.md and AGENTS.md are a symlink, two files or missing, and offers to create the symlink, with a copy fallback if symlinks do not work.
- Emits
scripts/sync-skills(Python 3, no network calls) with--checkfor CI (non-zero exit on drift) and--apply.
Who it is for
Developers who use Claude Code alongside Codex, OpenCode or GitHub Copilot and currently copy skills between folders by hand.
Example
"I keep my skills in .claude/skills and copy them to .codex by hand. Sync them, and give me a CI check."
You get a drift report (for example: 2 missing, 1 changed, 1 extra in .codex), a confirmation question before any write, the updated .codex copies, the sync script, and a list of files to commit.
What is included
SKILL.mdreferences/target-mapping.mdscripts/sync-skills
Requirements and limits
- Python 3; a git repo or plain folder. No network access.
- Supports only the four folders above, not other agents or future formats.
- The per-target mapping table is not verified against live docs; unverified cells are marked "check current docs" and should be confirmed. Unsupported-frontmatter detection stays off until you provide a verified field list.
- Dry run by default. Skills and the CLAUDE.md/AGENTS.md pair only: no settings, MCP configs or hooks, no description rewriting, no security scanning, no remote installs.
Requires Python 3. Limited to .claude/skills, .codex, .opencode, and .github folders. Does not scan for security vulnerabilities or install remote skills.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before any payment.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
Fresh listing
Recently published to Agensi
Money-back guarantee
Doesn't do what its page promises? You get your money back.
Trust & safety
Security scan
Security scan passed 12 days ago