bulkhead
v0.3.7by hashgraph-online · author: Myeongseok Kang
Guards for a machine where many coding agent sessions run at once: blocks name-based kills of another session's processes, port collisions, git commands that would move or delete another session's tree, whole-disk scans and long foreground sleeps, and holds heavy commands back when RAM runs low.
Package we checked
The manifest we fetched is plugins/dooly6276/claude-bulkhead/plugin.json in hashgraph-online/awesome-codex-plugins. The plugin root is the directory that holds that file. Its $schema is https://agent-plugins.org/schemas/1.0.0/plugin.schema.json. The manifest lists 1 skill (bulkhead) and no MCP servers.
What bulkhead does, in the publisher's words
claude-bulkhead
Stop Claude Code from running pkill node on your other dev servers, taking their ports, and committing another session's edits with git add -A.
Hooks that know which processes, ports and git trees belong to which Claude Code session and subagent, so they block only what would hurt something else, and tell Claude the safe command to run instead.
From the project README, punctuation lightly normalized · Open the README on GitHub
- Skills
- 1
- MCP servers
- 0
- Stars
- 1,568
- License
- MIT
- Repo created
- 2026-03-28
- Last pushed
- 2026-11-13
- Publisher type
- Organization
- Version
- 0.3.7
Links
Skills · 1
- bulkhead
Category
Other. Assigned by this directory. Agent Plugins 1.0.0 has no category field, so no manifest declares one.
Keywords
hooks · guard · processes · ports · memory · worktrees · git · parallel · agents · copilot-cli
Related by publisher and keywords
Reconstruct evidence-backed developer work from authorized local coding history and let Codex turn it into polished recaps.
Leeway is a drive for you and your agents. This plugin connects Codex to your Leeway drive at https://leeway.to/mcp and teaches it the drive's working rules: save without overwriting anyone's change, find what's there, undo from history, edit in a workspace and publish, and leave a hand-off note.
Long-term memory with no LLM in the loop: a local MCP server stores messages word for word in SQLite and recalls the ones that answer a question within a token budget.
Markdown-first long-term memory shared across your coding agents: one local memtomem store, searchable from Hermes, Claude Code, Codex CLI and Kimi Code.
Persistent memory for AI agents, shared across tools. Connects the hosted Mnemoverse memory server over remote MCP with a one-time browser sign-in: recall relevant memories before acting, save durable decisions and corrections, keep shared rooms across accounts. Bundles the CC0 agent-memory-discipline skill so the agent actually uses the memory it has.
Hand Rush a task: an agent does the work in its own cloud computer and returns a pull request, a report, files or an answer, with a record of what it did.
[](https://agentpluginsdirectory.com/plugins/bulkhead)✓ Verified . We fetched the manifest from GitHub and checked it against the official Agent Plugins 1.0.0 schema at agent-plugins.org.