by 1Password
Enables Codex to securely manage 1Password Developer Environments via the local MCP server, allowing listing environments, inspecting and adding variables, and creating .env mounts without exposing secret values.
The plugin connects Codex to the 1Password desktop app's local MCP server, providing AI‑driven access to Developer Environments while keeping secret values under 1Password control.
.agents/plugins/marketplace.json)..env files.scripts/package-skill-submission.sh builds a compliant ZIP for OpenAI’s plugin directory; validation script ensures correct structure..env files from stored secrets.Q: Do secret values ever get sent to Codex? A: No. The plugin only exposes variable names and metadata; actual secret values remain locked in the 1Password vault.
Q: Which operating systems are supported? A: macOS, Windows, and Linux.
Q: What if the “Integrate with MCP clients” setting is disabled? A: Ask an administrator to enable it under Policies → Sharing and permissions → Agentic permissions → Local MCP server.
Q: Can I submit this plugin to the OpenAI marketplace? A: Yes. Use the provided packaging script to create a “Skills Only” ZIP; the script strips MCP‑specific files to meet marketplace requirements.
Q: Is any additional server setup required?
A: No. The 1password-mcp binary is installed with the desktop app and runs automatically when Codex invokes the MCP tools.
This repository contains a Codex plugin that connects Codex to the 1Password desktop app's local MCP server for 1Password Developer Environments.
The plugin lets Codex help with secure project environment setup, such as
listing Developer Environments, inspecting variable names, adding variables, and
creating local .env mounts. Secret values remain controlled by 1Password and
are never shared with Codex.
Note: This plugin supports macOS, Windows, and Linux.
https://www.1password.dev/environments/mcp-codex-server
plugins/1password/.codex-plugin/plugin.json - Codex plugin metadata and UI
configuration.plugins/1password/.mcp.json - MCP server configuration for the 1Password
desktop app's bundled 1password-mcp binary.plugins/1password/skills/1password-environments/SKILL.md - Codex skill
instructions for using 1Password Developer Environments.plugins/1password/skills/1password-environments/agents/openai.yaml - Skill
interface settings (display name, icons, brand color). These belong here
rather than in SKILL.md frontmatter, which only carries the skill's name,
description, and other Agent Skills spec fields.plugins/1password/assets/ - Plugin icon and logo assets..agents/plugins/marketplace.json - Local marketplace entry that points Codex
at plugins/1password.scripts/package-skill-submission.sh - Builds the ZIP for a "Skills Only"
submission to the OpenAI plugin directory.scripts/validate_submission_zip.py - Checks a built ZIP against OpenAI's
submission rules.If Integrate with MCP clients is disabled by an administrator, ask your administrator to check Policies > Sharing and permissions > Agentic permissions > Local MCP server.
The MCP server runs via the 1password-mcp alias, which the 1Password
desktop app installs on your PATH:
1password-mcp
Once installed or enabled in Codex, you can ask for tasks like:
The 1Password desktop app may prompt for approval when Codex connects to the MCP server or accesses an Environment.
The plugin exposes these 1Password MCP tools to Codex:
authenticatelist_environmentscreate_environmentrename_environmentlist_variablesappend_variablescreate_local_env_filelist_local_env_filesTo build the ZIP for a Skills Only submission to the OpenAI plugin directory:
./scripts/package-skill-submission.sh
This writes dist/1password-skill-submission.zip and validates it before
exiting. Upload that file in the submission portal.
A skills-only bundle may not declare mcpServers, .mcp.json, apps,
.app.json, or interface.screenshots, but this repo keeps .mcp.json so the
plugin still works as a local install. The script resolves that by staging a
copy and stripping those pieces there, so your working tree is never modified.
Two structural rules cause most rejections, and the script handles both:
.codex-plugin/plugin.json
sits either at the ZIP root or inside its only top-level directory. Zipping
the repo root fails (several top-level entries), and so does zipping a bare
skill folder (no manifest).__MACOSX entries.To check an archive that was built some other way:
python3 scripts/validate_submission_zip.py dist/1password-skill-submission.zip
Note that a skills-only listing installs the skill but does not configure the MCP server, so users follow the Requirements section of SKILL.md to enable it themselves. Submitting the MCP server through the directory instead would require a production HTTPS server URL with domain verification, which does not apply to a local stdio server.
MIT. See LICENSE.
Please log in to share your review and rating for this MCP.
Explore related MCPs that share similar capabilities and solve comparable challenges
by chaitin
A self‑hosted web application firewall and reverse proxy that protects web applications from attacks and exploits by filtering, monitoring, and blocking malicious HTTP/S traffic.
by snyk
Scans installed AI agent components, MCP servers, and skill files for prompt‑injection, tool poisoning, toxic flows, hard‑coded secrets and other supply‑chain risks.
by OpenOSINT
Provides an AI‑driven OSINT workflow that lets users query a natural‑language REPL, CLI, web UI, or MCP server, automatically selecting and chaining 18 reconnaissance tools to collect, pivot, verify, and report public‑source intelligence.
by 2akouwu
Provides deterministic verification of AI‑generated claims against binary artifacts, ensuring every structural or behavioral assertion is grounded in the actual file and preserving verified state across context resets.
by hashgraph-online
Open-source antivirus for AI agents that blocks risky tools, secret exposure, prompt injection, malicious packages, MCP servers, plugins, and skills at runtime.
by safedep
Provides enterprise‑grade open source software supply chain security by scanning source code, dependencies, containers and SBOMs, detecting vulnerabilities and malicious packages, and enforcing policy as code.
by tufantunc
Provides controlled SSH access for LLM agents with command classification, policy‑based authorization, human‑in‑the‑loop approval, and immutable audit logging.
by semgrep
Offers an MCP server that lets LLMs, agents, and IDEs run Semgrep scans to detect security vulnerabilities in source code.
by KeyValueSoftwareSystems
Enables adversary emulation for AI agents, LLM applications, and MCP servers, letting teams test their AI systems against realistic attack scenarios.