Claude Code mod

protocol-guard

Required workflow checks with Claude Code function hooks. Refuses a knowledge-file write, or a memory-service save in external memory mode, until knowledge-save is open, refuses hand edits to generated indexes, refuses a pull request, close or merge until knowledge-save (and the work or merge-and-clean-up skill) is open, and holds a final reply once when working memory or the index rebuild and checker did not follow the change. Fact checks only: no model call and no reading of the agent's words. Runs only when CLAUDE_CODE_ENABLE_FUNCTION_HOOKS=1.

Guard & safety mods Plugin marketplace Source

Install

Inside Claude Code
/plugin marketplace add Mar5929/claude-toolkit
/plugin install protocol-guard@claude-toolkit

Mods install like plugins. Run /reload-plugins or start a new session to load it.

What it can access

Reach
L2Writes or runs: sets env vars, reads files, draws
Sees
compaction, every prompt, classic hooks, every skill, every tool call
Hooks
session.startsession.endturn.startsession.compactprompt.submitclassic.UserPromptSubmitclassic.Stopskill.prompttool.callturn.stepturn.complete

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.
Claude Code docs

Details

Repository
Mar5929/claude-toolkit
Folder
plugins/protocol-guard
Marketplace
claude-toolkit
Author
Mike Rihm
Type
Guard & safety mods
Last updated
26 September 2026
Listed in
awesome-claude-code-mods

FAQ

What is protocol-guard?

Required workflow checks with Claude Code function hooks. Refuses a knowledge-file write, or a memory-service save in external memory mode, until knowledge-save is open, refuses hand edits to generated indexes, refuses a pull request, close or merge until knowledge-save (and the work or merge-and-clean-up skill) is open, and holds a final reply once when working memory or the index rebuild and checker did not follow the change. Fact checks only: no model call and no reading of the agent's words. Runs only when CLAUDE_CODE_ENABLE_FUNCTION_HOOKS=1. It is a mod for Claude Code, from Mar5929/claude-toolkit on GitHub.

What can the protocol-guard mod access?

Its widest reach is "writes or runs" (sets env vars, reads files, draws). It sees compaction, every prompt, classic hooks, every skill, every tool call. Like any mod, it runs with your permissions.

How do I install protocol-guard?

Inside Claude Code (v2.1.287 or later), run /plugin marketplace add Mar5929/claude-toolkit and then /plugin install protocol-guard@claude-toolkit.

More guard & safety mods

prose-guard

cprentice9/ccmods

Blocks em dashes in edits, shell commands and replies, and British spellings in prose files, commit messages and pull requests.

Guard & safety mods

publish-guard

seanyyoo/claude-code-mods

Stops Claude from typing private terms into the browser or writing them into your publishing files, and keeps a log of what it blocked

Guard & safety mods/plugin install

prompt-redact

ezoushen/claude-mods

Replaces secrets and credentials with stable placeholders inside Claude Code before the model or the session file ever sees them — prompts, tool results and replies, plus attachments and compaction summaries where the platform lets this mod see them — and restores real values only in Edit/Write/NotebookEdit arguments. 220-rule corpus plus checksum detectors, a benign-shape allowlist, fail-closed scanning, a /redact pane and a redacted-values skill. No network, no subprocess, no model call.

Guard & safety mods/plugin install

push-guard

kk5190/claude-code-mods

Blocks git push and gh pr create until you confirm the commits, their authors and the diff size, with a side pane of the last check

Guard & safety mods/plugin install

port-guard

colorfield/claude-port-guard

Stops Claude from killing, or running E2E tests against, another project's local dev server

Guard & safety mods/plugin install

quarantine

ccdwyer/quarantine

Wraps web, MCP and third-party tool output as untrusted data and defangs prompt-injection lines before the model reads them

Guard & safety mods

Browse the most popular Claude Code mods, all guard & safety mods, or the skills and MCP server directories.