codex-multi-profile
Multi-workspace profile router, MCP server isolation auditor, and CLAUDE.md <-> AGENTS.md instruction bridge for Claude Code and OpenAI Codex.
No outside reports on file
This tool handles sensitive multi-profile separation and configuration routing, but its low visibility, unclear documentation, and unverified isolation claims make it risky for production use without thorough testing and code review.
Based on public adoption and transparency signals as of Oct 27, 2026. Our opinion, not a code audit or security guarantee.
What it can access
- Local filesNo local folder access notedLOW
- Command executionNo shell command runner notedLOW
- Network reachConnects to outside service APINOTED
- CredentialsNo API token requiredLOW
- Package sourcenpm (codex-multi-profile)PUBLIC
Inferred from this listing’s public package details and description. Check the project’s source before installing.
Install
npx -y codex-multi-profileAbout Codex Multi Profile
This MCP server acts as a router that manages multiple workspace profiles and claims to audit how different AI agents (like Claude Code and OpenAI Codex) are isolated from each other. It appears to bridge configuration files (CLAUDE.md and AGENTS.md) to control which instructions different AI tools follow.
Who it’s for
Developers who want to run multiple AI coding assistants in the same workspace while keeping their settings and access separate. Teams that need to audit or control what data different AI agents can see.
Alternatives
- direnv: Manages environment variables per directory; lightweight and well-tested.
- Claude Desktop native multi-profile support: Check if Claude's built-in workspace features meet your needs.
- Manual profile separation: Keep workspace folders in separate directories with separate tool configurations.
- Docker containers: Isolate each AI agent in its own container for stronger separation.
Setup
Check these steps against the project’s own instructions before running anything.
Follow the project's README. The package is published on npm as codex-multi-profile, but without clear documentation we recommend checking the GitHub repository for installation and configuration steps before use.
Why this grade
Signal-based grade (2/7 points): heavily downloaded, no public source code found. This reflects adoption and transparency, not a code review.
Popularity and upkeep
- Weekly npm downloads
- 255k
- Listing last refreshed
- 2026-10-28
These are public popularity signals, not a maintenance check. Look at the repository’s commit history and open issues.
Security reports
No reports found for this package in our research (last checked 2026-10-29).
Risks to watch for
- Access to multiple workspaces: This tool routes traffic between AI agents and multiple profiles, meaning it can see file paths, code, and settings across your project workspace.
- Configuration file control: It reads CLAUDE.md and AGENTS.md files that could contain secrets or instructions; an attacker could hide malicious rules in these files that an AI agent might follow without noticing.
- Low adoption signals: Zero GitHub stars and no Hacker News discussion means few eyes have reviewed this code for security issues.
- Unclear isolation claims: The 'auditor' and 'isolation' features are not well-documented, so it is hard to verify that separation between agents actually works as intended.
- Data leakage between profiles: If isolation fails, one AI agent's workspace profile could leak files or secrets meant only for another agent.
How to lower the risk
Least-privilege permissions
- Only grant this server access to workspace folders it truly needs to route between.
- Never give it access to your entire home directory or sensitive credential stores.
Network isolation
- Run it on a local machine only; do not expose it over the network.
- Verify it does not make outbound calls to unknown servers.
Monitoring
- Log which files and profiles this server reads and writes.
- Watch for suspicious instructions appearing in your CLAUDE.md or AGENTS.md files.
Red flags to watch for
- If configuration files appear without your edits, stop using it immediately.
- If one agent suddenly has access to another agent's workspace, the isolation may have failed.
Questions
How do I install Codex Multi Profile?
Run npx -y codex-multi-profile in your terminal, or add the server entry to your claude_desktop_config.json or .cursor/mcp.json as shown in the Setup section on this page.
Are there security reports about Codex Multi Profile?
No reports found for this package in our research (last checked 2026-10-29). Codex Multi Profile currently holds Grade C (Moderate risk signals) based on public adoption and transparency signals. Key risk areas to watch include access to multiple workspaces, configuration file control, low adoption signals.
Is Codex Multi Profile actively maintained?
Codex Multi Profile has 255k weekly npm downloads. Our directory data was last refreshed on 2026-10-28. Check recent commit activity, release tags, and open issues before pinning it in production.
What are alternatives to Codex Multi Profile?
Notable alternatives to Codex Multi Profile include direnv (Manages environment variables per directory), Claude Desktop native multi-profile support (Check if Claude's built-in workspace features meet your needs), Manual profile separation (Keep workspace folders in separate directories with separate tool configurations), Docker containers (Isolate each AI agent in its own container for stronger separation). You can also compare Codex Multi Profile side by side with other MCP servers in the Related tools section on this page.
README badge and team notes
Maintain or use Codex Multi Profile? Show its live grade in your GitHub README, docs site, or run a check from the terminal before you install.
GitHub README Markdown:
[](https://agentarmorhq.com/skills/hung2124-codex-multi-profile)HTML embed:
<a href="https://agentarmorhq.com/skills/hung2124-codex-multi-profile"><img src="https://agentarmorhq.com/api/badge/hung2124-codex-multi-profile" alt="AgentArmorHQ Risk Grade for Codex Multi Profile" height="20" /></a>Terminal check:
curl -fsSL https://agentarmorhq.com/api/cli | node --input-type=module - check "hung2124-codex-multi-profile"- Compare Codex Multi Profile →
- Add to a team allowlist
- Claude and terminal guide →
- Download the PDF guide
Share with your security lead
Need sign-off to use codex-multi-profile? Copy this note for Slack, a ticket or a pull request.
### AgentArmorHQ Security Review Memo: codex-multi-profile
- **Listing URL:** https://agentarmorhq.com/skills/hung2124-codex-multi-profile
- **Type:** App connector
- **Risk Grade:** Grade C (Moderate risk signals)
- **Outside Security Advisories on File:** 0
#### Permission & Blast-Radius Signals
- **Local files:** No local folder access noted
- **Command execution:** No shell command runner noted
- **Network reach:** Connects to outside service API
- **Credentials:** No API token required
- **Package source:** npm (codex-multi-profile)
#### Recommended Hardened Setup
- **Command:** `npx -y --ignore-scripts codex-multi-profile`
- **Controls:** Pin an exact version in `.mcp.json`, scope access to a single workspace folder, and use a least-privilege read-only token where supported.
_Based on public adoption and transparency signals as of Oct 27, 2026. Our opinion, not a code audit or security guarantee._Key facts
- Type
- App connector
- Install command
- npm install codex-multi-profile
- Est. context load
- ~550 tokens / turn (~3 tool schemas, 0.27% of 200k context)
- Popularity
- 255k weekly npm downloads
- Last updated
- 2026-10-28
- Last graded
- 2026-10-27
Community
Discussions for Codex Multi Profile
Ask questions or share configuration tips. Governed by our Community Rules & Terms.
No posts match this filter yet.
Be the first to ask a question or share a working tip.
Check before you install, in Claude
Your assistant answers Block, Review or No known issues for any package before it installs it.
Get this check inside Claude