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.
/plugin marketplace add felipearomani/claude-shipyardnpx agentmods add plugins/felipearomani/claude-shipyard/fleetgit clone --depth 1 https://github.com/felipearomani/claude-shipyardGrade A, and why
fleet 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 yesterday.
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.
What it actually says
{
"name": "fleet",
"version": "0.2.0",
"description": "Plan, dispatch and land work across a fleet of autonomous Claude Code agents.",
"author": {
"name": "Felipe Romani",
"url": "https://github.com/felipearomani"
},
"homepage": "https://github.com/felipearomani/claude-shipyard",
"repository": "https://github.com/felipearomani/claude-shipyard",
"license": "MIT",
"keywords": [
"multi-agent",
"orchestration",
"planning",
"autonomous-agents",
"code-review",
"fleet"
]
}
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.
- yesterday First seen · 21 lines scan A aaa755006de1
fleet is a plugin published in the GitHub repository felipearomani/claude-shipyard (1 stars, last pushed 5d ago), licensed MIT. Its token cost is not measured: this kind of file is read by the harness, not the model. 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 plugins, from other repositories
dos-kernel
The kernel is the part that doesn't believe the agents. Bundles the DOS hooks (verify-on-stop, deny-a-refused-call, re-surface-a-stalled-stream), the DOS MCP server (verify / arbitrate / refuse-with-a-reason as tools), and the generic skill pack — one install. Requires the dos-kernel Python package: pip install 'dos.
dos
DOS — the domain-free trust substrate for fleets of autonomous agents.
teamctl
teamctl — persistent AI agent teams, declared in YAML and launched from a chat with Claude.
wiff
Deterministic, resumable multi-agent workflows that orchestrate Codex, Claude, Cursor, and Kimi agents — five MCP tools plus the workflow-authoring skill.
wiff
Plugin marketplace listing 1 plugin: wiff.
embarak
Mohamed Embarak's Claude Code plugins.