Claude Code mod
agent-guard
Guard for dispatching subagents: denies isolation "worktree", checks a declared WORKTREE is a separate, clean, never-dispatched git worktree, and in opted-in repos requires non-read-only subagents to declare one.
Install
/plugin marketplace add AbyssCN/claude-lead-harness
/plugin install agent-guard@claude-lead-harnessMods install like plugins. Run /reload-plugins or start a new session to load it.
What it can access
- Reach
- L2Writes or runs: runs processes, reads files, persists state
- Sees
- Agent calls
- Hooks
tool.call
From a static scan of the mod's code by awesome-claude-code-mods. It shows what the code asks for, not a security review.
How Claude Code mods work
- A mod is a plugin with a JavaScript or TypeScript module that runs inside Claude Code (version 2.1.287 or later). It can draw panes beside the transcript, a band above the prompt, buttons and toasts, restyle the spinner and tool rows, and watch or rewrite tool calls and prompts.
- It installs like any plugin: add its marketplace with /plugin marketplace add owner/repo, then /plugin install name@marketplace.
- Check what a mod can reach before installing: it runs with your permissions. Reach goes from drawing only, to reading files, to writing files or running processes, to the network. Run claude plugin validate on a clone to list its hooks and calls without running it.
- Turn one off from the Installed tab in /plugin, or start Claude Code with --safe-mode to skip every customization for a session.
Details
- Repository
- AbyssCN/claude-lead-harness
- Folder
mods/agent-guard- Marketplace
claude-lead-harness- License
- MIT
- Last updated
- 10 October 2026
- Listed in
- awesome-claude-code-mods
FAQ
What is agent-guard?
Guard for dispatching subagents: denies isolation "worktree", checks a declared WORKTREE is a separate, clean, never-dispatched git worktree, and in opted-in repos requires non-read-only subagents to declare one. It is a mod for Claude Code, from AbyssCN/claude-lead-harness on GitHub.
What can the agent-guard mod access?
Its widest reach is "writes or runs" (runs processes, reads files, persists state). It sees Agent calls. Like any mod, it runs with your permissions.
How do I install agent-guard?
Inside Claude Code (v2.1.287 or later), run /plugin marketplace add AbyssCN/claude-lead-harness and then /plugin install agent-guard@claude-lead-harness.
More guard & safety mods
agent-cross-check
naoanao/agent-cross-check
When another coding agent (or a person) commits to your repo, Claude notices, runs mechanical checks (line endings, secrets, .env files, and the agent's report against the real files: missing files, silent videos, unconfirmed claims) and audits the commits by diff and tests. A git push waits until they are reviewed.
agent-state
kleinheinczg/claude-code-agent-state-mod
Reports what a Claude Code session is really doing (idle / working / tool / permission, plus signals and token use) from inside the session, as a JSON snapshot and a JSONL event log
adaflow-guard
Adalink-ai/ada-mods
Guardrail para apps integrados ao Adaflow: impede vazar o app token (x-ada-token) para o browser, para o codigo-fonte ou para o transcript
agent-state
M-i-k-e-l/agent-state
Shows the state of Claude (working, permission, input, idle, background, compacting) in the iTerm2 tab subtitle and in a Sessions pane in the desktop app.
ad-ldap
seanGSISG/ad-ldap-plugin
Active Directory administration over LDAPS for Claude Code: an MCP server (FastMCP + ldap3) with 15 tools, admin agents, slash commands, a safety skill, and a guard that refuses any AD write until the identical call has been dry-run first. Headline workflow: mass-assign computer managedBy from description-held owner names via a dry-run plan/review/apply loop.
approve-blocked
adammhal/claude-mods
Retries an auto-mode-blocked call up to three times so Claude Code's own permission prompt (and Moshi) can ask you.
Browse the most popular Claude Code mods, all guard & safety mods, or the skills and MCP server directories.