
Argent for OpenAI Codex CLI
Codex model traffic and tool actions pass through Argent, attributed to an agent your developer owns. Here is what Argent governs, how it connects and what your team sets.
What Argent does with OpenAI Codex CLI
Every call OpenAI Codex CLI makes to a model goes through Argent first. Argent knows which agent sent it and which person owns that agent, checks it against your policy, and writes a signed record of what it decided.
- A named agent with a human owner. Each install of OpenAI Codex CLI gets its own Argent agent identity tied to an accountable person, so no call is anonymous. Identity tethering
- Only the models you allow. The agent can reach only the providers and models your policy names, and a request can be swapped to an approved model. Model control
- Sensitive data handled before it leaves. Prompts and replies are scanned for personal and secret data, then redacted or blocked as your policy says. Sensitive data redaction
- Spend priced and capped. Every request is priced and metered against budgets for the agent, its owner and the teams above them. AI spend caps
- A signed record you can replay. Each decision becomes a hash-chained, signed audit record, and requests are grouped into sessions you can replay. Session replay
- Tool actions checked before they run. Shell commands, file reads and writes, web fetches, MCP calls and prompt submits ask Argent first; if Argent cannot answer, the action is refused. Agent permissions
How OpenAI Codex CLI connects
OpenAI Codex CLI reaches Argent in two ways.
- TLS-inspecting forward proxy. OpenAI Codex CLI keeps its normal provider settings and sends its HTTPS traffic through the Argent forward proxy and trusts your Argent certificate; the exact settings are below. The proxy signs in with the agent's own key, so every call is attributed, then runs the same checks as the gateway. It opens only known AI hosts, and its certificate is limited to those hosts.
- Tool hooks. Before OpenAI Codex CLI runs a tool or submits a prompt, its hook asks Argent for a decision and a data-loss scan. If Argent cannot answer, the hook refuses.
One command does the setup. argent connect codex registers a per-tool agent under the signed-in person, keeps its key only in the OS keychain, and writes the tool's own config with a backup. --revert puts the old config back. Keyless agents explains why keys stay out of config files.
A small loopback relay holds the agent key and adds it to each proxy connection, so the key is never written into the tool's settings.
Proven with the real tool. Argent's team has driven OpenAI Codex CLI itself through a live Argent gateway and checked the signed records it produced.
The same steps are in the Argent console under Integrations > AI Clients, and each agent's Connect tab gives the copy-paste values for your tenant.
What you configure
What your team sets in OpenAI Codex CLI. These are the settings the Argent console shows for this tool. Values in angle brackets come from your tenant.
- Sign in and connect.
argent login, thenargent connect codex. Writes ~/.codex/config.toml: model_provider="argent"; [model_providers.argent] name="Argent", base_url="https://api.openai.com/v1" (traffic reaches Argent via the forward proxy), wire_api="responses"; [model_providers.argent.auth] command=<argent bin>, args=["key","codex"], timeout_ms=5000; plus hooks and OTLP export. - Manual proxy form. Run codex with HTTPS_PROXY=<http:// loopback relay to the Argent forward proxy> and CODEX_CA_CERTIFICATE=<per-tenant intercept CA PEM> (fallback SSL_CERT_FILE). Set before login.
- Alternative: gateway base URL. openai_base_url = "https://<argent-gateway>/<tenant>/openai/v1" in ~/.codex/config.toml, or a custom [model_providers.<id>] with base_url, env_key, wire_api="responses". The OPENAI_BASE_URL env var fallback was removed in April 2026; use the TOML key.
- Undo.
argent disconnect codex.
Roll it out to every machine
Make it the only way OpenAI Codex CLI runs on company machines. These controls are documented by the platform each one belongs to (linked); pair them with the Argent settings above.
- requirements.toml (admin-enforced, cannot be overridden by user config or session choices) (macOS). Pin model_provider and a complete [model_providers.<id>] definition (including base_url) so Codex can only talk to the Argent-routed provider; also allowed_approval_policies, allowed_sandbox_modes, MCP server allowlist, rules/prefix_rules (decision prompt|forbidden only). Vendor docs
- requirements.toml (Windows). Same as above. Vendor docs
- ChatGPT Enterprise workspace-managed requirements (cloud). Admin-enforced requirements delivered to signed-in local Codex clients of the workspace. Vendor docs
- Environment for proxy and CA (deploy via the same OS tooling). HTTPS_PROXY/NO_PROXY and CODEX_CA_CERTIFICATE (or SSL_CERT_FILE) for the Codex process. These are environment, not requirements.toml keys; shell_environment_policy only governs what spawned commands inherit. Vendor docs
Argent ships the push kit. For managed fleets the console produces per-MDM scripts (Jamf, Kandji, Intune-style, shell and PowerShell) with your tenant certificate and proxy address built in, plus a fleet setup script and certificate bundle. Your company CA can sign the Argent certificate, so managed machines already trust it.
Want to see which AI tools people already run before you push anything? Start with Shadow AI discovery in watch mode. Watch before you enforce
Check that it works
Send one request and look for the signed row. argent doctor codex, run codex exec 'say OK', confirm a signed llm ALLOW row with metering for the argent-codex-<host> agent in the Argent console live stream.
Every request also appears on the live stream with the agent's name and owner, and in the session view. Session replay
Where to go next
Keep going. Pages that cover the same ground from another angle.
- Use case: Securing AI coding agents
- Goals: See all AI in your company, Data loss prevention, Budget and token spend, Model control and routing, Agent identity and ownership
- Capabilities: Keyless agents, Prompt injection defense, Response screening, One-click audit evidence
- Related integrations: Claude Code, Gemini CLI, GitHub Copilot CLI, Cursor Agent CLI
- All integrations: Browse every AI client, provider, vault and identity source
Want this running on your own use cases?
Book a demo