Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add BreetyGreen/Myco --skill multi-agent-skill-sharinggit clone --depth 1 https://github.com/BreetyGreen/MycoWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/breetygreen/myco/multi-agent-skill-sharing)<a href="https://agentmods.dev/skills/breetygreen/myco/multi-agent-skill-sharing"><img src="https://agentmods.dev/badge/skills/breetygreen/myco/multi-agent-skill-sharing/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/breetygreen/myco/multi-agent-skill-sharing"><img src="https://agentmods.dev/badge/skills/breetygreen/myco/multi-agent-skill-sharing.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00141 | $0.02471 |
| Opus 5.5 | $0.00056 | $0.00988 |
| Sonnet 5.5 | $0.00028 | $0.00494 |
| Haiku 4.5 | $0.00014 | $0.00247 |
Grade A, and why
multi-agent-skill-sharing scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2mo ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 207 lines — stays where its author put it; the contents beside it link to each section on GitHub.
multi-agent-skill-sharing
Install a skill once and make every AI coding agent working on the same repository able to read it and invoke it in its own way.
This skill is tool-agnostic. It works whether you use two agents or six.
When to use
- You use multiple AI coding products on the same project (e.g. Claude Code
- Codex + Cursor) and want a skill / ruleset / prompt to be shared and switchable between them.
- Someone asks "why can only one tool use this skill?" or "why can't I call it in tool X?"
- A skill is installed but one agent can't find it.
- You want a skill to live inside the repo and travel with Git, so it survives a new machine or a new teammate.
The core mental model (read this first, or you'll install it wrong)
There is no shared skills directory across agents. "Install once, all tools see it" is literally false. Each product reads skills from a different directory, and the invocation syntax differs too.
Real sharing is achieved by installing the skill into per-agent directories that live inside the repository, so it travels with Git. One physical copy per agent, all committed, all in sync.
Where each agent reads skills + how you invoke them
⚠️ Verify paths against the current version of each tool before relying on this table. These conventions move fast. Last verified: 2026-07 (see "Verification" below). When in doubt, check the tool's own docs or probe the filesystem.
| Agent | Repo-level dir (travels with Git) | User-level dir | How to invoke |
|---|---|---|---|
| Claude Code | <repo>/.claude/skills/ |
~/.claude/skills/ |
Mention the skill by name in chat. Some suites (e.g. gstack) add /slash commands on top. |
| WorkBuddy | <repo>/.workbuddy/skills/ |
~/.workbuddy/skills/ |
Mention the skill by name (same convention as Claude Code). |
| Codex CLI | <repo>/.agents/skills/ and/or <repo>/.codex/skills/ (see note) |
~/.agents/skills/ or ~/.codex/skills/ (see note) |
$skill-name, the /skills picker, or just name it in the prompt. NOT a /design-style slash command. |
| Cursor | <repo>/.cursor/skills/ (also reads .cursor/rules/ .mdc) |
— | Rules auto-inject; also tolerates .agents/ skills. |
| Antigravity | <repo>/.antigravity/skills/ (VS Code–based, convention) |
— | Mention the skill by name; vendor path unverified. |
| Gemini CLI | <repo>/.agents/skills/ |
— | Auto-discovered; name it in the prompt. |
| Cline | <repo>/.cline/skills/, .clinerules/skills/, or .claude/skills/ |
~/.cline/skills/ |
Name it; gated behind an experimental toggle. |
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 2mo ago First seen · 207 lines · 141 tokens per session scan A b2c4dc8a3ea5
multi-agent-skill-sharing is a skill published in the GitHub repository BreetyGreen/Myco (2 stars, last pushed 3mo ago), licensed MIT. It adds 141 tokens to every session and 2,471 once invoked, about $0.0006 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
agent-team-protocol
Multi-agent delegation rules, three-phase workflow (Plan/Build/Review), model selection, collaboration gates, and Ralph Loop integration.
opus-4-7-features
Optional guidance for sessions using Claude Opus 4.7: adaptive thinking, xhigh effort, prompt caching, /btw side questions.
phase-workflow
Phase transition rules, post-phase verification steps, peer review validation, and commit-gate approval flow.
x-spec
A system-planning guide that turns a vague idea into a set of linked design documents. It defines goals, modules, interfaces, data, workflows, and ways to check the result.
copilot-conventions
Cross-copilot portable conventions: target alignment, complexity control, minimal changes, git discipline, and project structure.
integration-testing
Test integration points early. Verify cross-service contracts, API boundaries, and data flow before declaring done.