agentleFS
Sign inStart free
Martin-HausleitnerMartin-Hausleitner/martins-awesome-skills/skills/software-development/cross-agent-skill-sync/SKILL.md

cross-agent-skill-sync

Skill3 starsChanged 6 months agoMITView on GitHub

Use when discovering, verifying, or sharing SKILL.md based skills across local Hermes, Codex, OpenClaw, Gemini, OpenCode, Claude-adjacent, and generic agent skill roots without overwriting existing skills or leaking private paths.

1 other project uses a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

8 rules. About 0 are common, 3 are distinctive, and 5 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈3

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 5

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Use this skill when the user wants the same skill layer to be visible to multiple local agents, including Hermes, Codex, OpenClaw-style …

    line 15
  2. Distinctive:

    Never overwrite an existing skill unless the user explicitly asks for a destructive replacement workflow.

    line 25
  3. Very specific:

    Prefer --strategy symlink for local same-machine sharing.

    line 26
  4. Very specific:

    Treat paths, prompts, profiles, tokens, memory, browser state, and agent logs as private.

    line 27
  5. Very specific:

    Use --execute only after the dry-run plan looks correct.

    line 28
  6. Very specific:

    Use --require-skill <name> for must-have skills, especially bridge skills such as codex-computer-use-eu-activate.

    line 29
  7. Distinctive:

    Add an external local skill, such as a Codex Computer Use activation bridge, without committing its private path:

    line 43
  8. Distinctive:

    Run the deterministic tests:

    line 64

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Cross-Agent Skill Sync

Use this skill when the user wants the same skill layer to be visible to multiple local agents, including Hermes, Codex, OpenClaw-style workspaces, Gemini CLI, OpenCode, Claude-adjacent setups, and generic SKILL.md consumers.

The helper is additive and dry-run-first:

node skills/software-development/cross-agent-skill-sync/scripts/cross_agent_skill_sync.mjs --target all

Operating Rules

  • Never overwrite an existing skill unless the user explicitly asks for a destructive replacement workflow.
  • Prefer --strategy symlink for local same-machine sharing; use --strategy copy for portable snapshots or remote machines.
  • Treat paths, prompts, profiles, tokens, memory, browser state, and agent logs as private. The report redacts $HOME by default.
  • Use --execute only after the dry-run plan looks correct.
  • Use --require-skill <name> for must-have skills, especially bridge skills such as codex-computer-use-eu-activate.
  • For Claude Code, project skills only into a configured local skill directory or emit docs/indexes. Do not assume Claude consumes arbitrary SKILL.md folders unless the local setup has that bridge.

Common Local Sync

Plan a sync of the repository skills into all known local agent roots:

node skills/software-development/cross-agent-skill-sync/scripts/cross_agent_skill_sync.mjs \
  --target all \
  --require-skill ai-research-browser \
  --require-skill oracle-ai-research-e2e

Add an external local skill, such as a Codex Computer Use activation bridge, without committing its private path:

node skills/software-development/cross-agent-skill-sync/scripts/cross_agent_skill_sync.mjs \
  --target hermes \
  --include-skill /path/to/codex-computer-use-eu-activate/skill/codex-computer-use-eu-activate \
  --require-skill codex-computer-use-eu-activate

Execute after review:

node skills/software-development/cross-agent-skill-sync/scripts/cross_agent_skill_sync.mjs \
  --target all \
  --strategy symlink \
  --execute \
  --out /tmp/cross-agent-skill-sync-report.json

Verification

Run the deterministic tests:

node --test skills/software-development/cross-agent-skill-sync/tests/cross_agent_skill_sync.test.mjs

Then run the repository-wide public safety audit:

scripts/audit-public-safety.sh

Status Meanings

  • install: destination is missing and will be created when --execute is set.
  • installed: destination was created in this run.
  • exists: destination already exists and was left untouched.
  • conflict: destination exists with different content or an unexpected shape; manual review needed.
  • missing-required-skill: a required skill was not found in the selected sources.

Good Boundaries

This skill shares instructions and helper files. It does not share browser cookies, provider sessions, private memories, shell histories, token files, or agent transcripts.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

Check your setup, free